Download pdf - Further contributions towards the molecular analysis of … · FCUP Further contributions towards the molecular analysis of NIPBL and SMC1A genes in a cohort of patients with Cornelia

FCUP Further contributions towards the molecular analysis of NIPBL and SMC1A genes in a cohort of

patients with Cornelia de Lange Syndrome

Agradecimentos

Quero agradecer a todos aqueles que me apoiaram e que de alguma forma

contribuíram para a realização deste trabalho: aos meus colegas da Unidade de

Genética Molecular. Ao professor Doutor Mário Sousa pela sua orientação e

disponibilidade, à Dra. Rosário Santos por me ter possibilitado a realização da minha

tese na Unidade de Genética Molecular, e especialmente ao Dr. Jorge Oliveira por

todo o seu apoio, conselhos e disponibilidade.



ABSTRACT:

Cornelia de Lange Syndrome [CdLS (MIM#122470)] is a rare multisystemic disorder,

characterized by a typical phenotype that includes distinctive facial dysmorphism,

hirsutism, growth and psychomotor developmental delay, limb defects, multiple organ

system problems such as frequent gastrointestinal and congenital heart defects. CdLS

is essentially caused by defects in NIPBL and SMC1A genes (~50% and 5% of cases,

respectively), but less frequently mutations have been also described in other genes

(SMC3, HDCA8, RAD21). This genetic heterogeneity in CdLS is partially explained by

a close functional relationship at the cellular level, since all these genes encode for

proteins involved in the sister chromatid cohesion complex. The molecular and clinical

characterization of CdLS patients was initiated at the national level in 2005. This initial

part of the work enabled the detailed characterization of thirteen CdLS patients with

novel NIPBL mutations, and the development of a locus specific database for this gene

(Oliveira et al., 2010).

The present work intended to unfold new molecular findings in the remaining 40 CdLS

patients of our cohort. Molecularly unresolved patients were also analyzed by other

techniques such as multiplex ligation-dependent probe amplification technique for

NIPBL gene, and by high resolution melting curve analysis (hrMCA) for SMC1A. The

analysis of NIPBL gene identified eight mutations previously reported in the literature

(including a case with suspected somatic mosaicism), three novel mutations (c.86del,

c.6983C>G and c.7307C>T) and two large deletions. The development of hrMCA

applied to SMC1A is a fast and cost-effective scanning method, and allowed the

identification of one mutation (c.1487G>A).

Overall, mutations in SMC1A and NIPBL have been identified in 51% of the

Portuguese CdLS patients studied by our group. The combination of different

experimental procedure was essential for attaining an enhanced mutation detection

rate in CdLS. Considering the distinct gene mutation detection rates a flowchart is

proposed for the genetic molecular diagnostic of CdLS. Due to recent discovery of new

genes involved in the disease, and considering that several patients are still

molecularly uncharacterized, its plausible that in a near future new genetics causes for

CdLS will be identified.

Keywords: Cornelia de Lange, CdLS, NIPBL, SMC1A, hrMCA, MLPA, Cohesin,

Mutation screening



RESUMO:

A Síndrome Cornélia de Lange [CdLS (MIM#122470)] é uma doença multissistémica

rara, caracterizada por um fenótipo típico que inclui um dismorfismo facial

característico, hirsutismo, atraso no crescimento e desenvolvimento psicomotor,

defeitos nos membros, problemas associados a múltiplos sistemas de órgãos, sendo

os mais frequentes problemas cardíacos, e refluxo gastrointestinal. A CdLS está

associada a defeitos nos genes NIPBL e SMC1A genes (identificados em ~50% e 5%

dos doentes, respectivamente), mas outras mutações mais raras foram descritos em

outros genes (SMC3, HDCA8, RAD21). A heterogeneidade genética nesta síndrome é

explicada ao nível celular, pela proximidade funcional das proteínas codificadas por

estes genes, estando envolvidas no complexo das coesinas que unem os cromatídeos

irmãos. Em 2005, iniciou-se em Portugal a caracterização clínica e molecular dos

doentes com CdLS. A parte inicial deste projecto permitiu a descrição detalhada de

treze doentes com novas mutações no gene NIPBL e o desenvolvimento de uma base

de dados de mutações específica para este locus (Oliveira et al., 2010).

Com a realização desta dissertação de mestrado pretende-se contribuir na

caracterização molecular dos restantes 40 doentes Portugueses do grupo de estudo.

Os casos não esclarecidos molecularmente foram analisados por outras técnicas

nomeadamente, multiplex ligation-dependent probe amplification technique aplicada ao

gene NIPBL, e por high resolution melting curve analysis (hrMCA) desenvolvida para o

gene SMC1A. A análise do gene NIPBL permitiu a identificação de oito mutações

anteriormente descritas na literatura (incluindo um caso com suspeita de mosaicismo

somático), três novas mutações (c.86del, c.6983C>G e c.7307C>T), bem como duas

grandes delecções neste gene. O desenvolvimento da técnica hrMCA aplicada ao

gene SMC1A demonstrou ser eficaz, tratando-se de um método rápido e rentável,

tendo permitido a identificação de uma mutação (c.1487G>A).

Considerando todos os dados obtidos, foram identificadas mutações em 51% dos

doentes Portugueses com CdLS. A utilização de diferentes técnicas moleculares foi

essencial para atingir uma taxa de detecção de mutações mais elevada. Devido à

recente descoberta de dois genes envolvidos na doença, (HDCA8, RAD21), e

existindo ainda uma percentagem razoável de casos sem mutação identificada, é

possível que novas causas genéticas de CdLS sejam identificadas neste doentes.

Palavras-Chave: Cornelia de Lange, CdLS, NIPBL, SMC1A, Coesina, hrMCA, MLPA,

rastreio de mutações



Index

1. Introduction ..................................................................................................... 1

1.1. Cornelia de Lange Syndrome .................................................................. 1

1.1.1. History and Nomenclature ................................................................ 1

1.1.2. Clinical features ................................................................................ 2

1.1.3. Etiology............................................................................................. 3

1.2. Biochemistry basis of CdLS - the cohesin complex .................................. 3

1.3. Genetics of CdLS .................................................................................... 6

1.3.1. NIPBL gene ...................................................................................... 6

1.3.2. SMC1A gene .................................................................................... 7

1.3.3. SMC3 gene ...................................................................................... 9

1.3.4. HDAC8 gene .................................................................................... 9

2. Objectives ..................................................................................................... 10

3. Materials and Methods ................................................................................. 11

3.1. Patients ................................................................................................. 11

3.2. Preparation of biological samples .......................................................... 12

3.2.1. DNA extraction method ................................................................... 12

3.2.2. RNA extraction method ................................................................... 12

3.2.3. Nucleic acid quantification .............................................................. 12

3.2.4. RNA conversion to cDNA ............................................................... 13

3.2.5. Aliquots preparation ........................................................................ 13

3.3. DNA amplification by PCR ..................................................................... 13

3.3.1. Oligonucleotides design .................................................................. 14

3.3.2. PCR mixture and thermocycling profiles ......................................... 15

3.4. Agarose gel electrophoresis .................................................................. 17

3.5. PCR purification ..................................................................................... 18

3.6. Sanger Sequencing ............................................................................... 18

3.6.1. Sequencing reaction ....................................................................... 18



3.6.2. Sequencing PCR purification .......................................................... 19

3.6.3. Sequencing analysis ....................................................................... 19

3.7. Multiplex ligation probe amplification ..................................................... 19

3.8. High Resolution Melting Curve Analysis ................................................ 20

3.9. Bioinformatics and databases ................................................................ 22

3.9.1. Missense mutations analysis .......................................................... 22

3.9.2. Algorithms for evaluating splicing.................................................... 22

4. Results ......................................................................................................... 24

4.1. NIPBL gene sequencing ........................................................................ 24

4.1.1. Mutations causing premature stop codons. ..................................... 27

4.1.2. In-frame deletion ............................................................................. 28

4.1.3. Missense mutations ........................................................................ 28

4.1.4. Splicing mutations .......................................................................... 31

4.1.5. Variant with unknown significance .................................................. 32

4.2. Large mutations detected in NIPBL gene ............................................... 33

4.3. Mutation screening for SMC1A gene ..................................................... 34

5. Discussion .................................................................................................... 38

5.1. Mutations identified in NIPBL ................................................................. 38

5.1.1. Detection of large NIPBL deletions by MLPA .................................. 40

5.1.2. Somatic mosaicism ......................................................................... 41

5.1.3. Novel silent polymorphism .............................................................. 42

5.2. Implementation of hrMCA for SMC1A .................................................... 42

5.3. Mutation detection rate .......................................................................... 43

5.4. Bioinformatic tools and databases ......................................................... 44

5.5. Procedure for the molecular diagnostic of CdLS patients ....................... 45

5.6. Future perspectives ............................................................................... 47

6. Bibliography .................................................................................................. 48



Tables and Figures Index

Figure 1.1 – Schematic representation of the cohesin complex……………………4

Table 1.1- Members and regulators of the cohesin complex……………………….5

Figure 1.2 – NIPBL-LOVD database content analysis………………………………7

Table 3.1: PCR program for NIPBL amplification…………………………………..16

Table 3.2: Touchdown PCR program for NIPBL exon 44…………………………17

Table 3.3: High Resolution Melting Curve Analysis for SMC1A gene……………21

Figure 4.1: Screening for NIPBL gene mutations by sequencing…………………24

Table 4.1: Mutations found in NIPBL gene………………………………………………...25

Figure 4.2: Representation of the NIPBL gene……………………………………………26

Figure 4.3: Sequencing electropherogram of a novel small deletion……………….…..27

Figure 4.4: Suspected mosaicism mutation c.1885C>T…………………………….……28

Figure 4.5: Sequencing electropherograms of two missense mutations………….……29

Table 4.2: Evaluation of missense variant in terms of pathogenicity……………………30

Figure 4.6: Protein sequence alignments of delagin from several organisms………….30

Figure 4.7: Evaluation of P46 splicing mutation c.64+1G>A……………………………..31

Figure 4.8: Evaluation of patient P20 splicing mutation c.6763+5G>T…………………32

Figure 4.9: Variation detected in patient P44………………………………………………33

Figure 4.10: cDNA electropherogram analysis……………………………………………33

Figure 4.11: Screening of NIPBL mutations by MLPA……………………………………33

Figure 4.12: MLPA results of patient P32………………………………………………….34

Figure 4.13: Strategy for SMC1A mutation screening by hrMCA………………………35 Figure 4.14: Normalized High Resolut ion Melting Curves, for amplicons of exons 1 to 6a……………………………………………………………………………………………….35

Figure 4.15: Normalized High Resolution Melting Curves, for amplicons of exons 6b to 25………………………………………………………………………………………………36

Figure 4.16: SMC1A missense mutation, detected through High Resolution Melting Curve Analysis………………………………………………………………………….........37

Figure 5.1: Representation of the NIPBL gene, with the distribution of all known point and small del/ins mutations throughout the gene…………………………………………39

Table 5.1: Overall results of the genetic analysis of NIPBL and SMC1A genes in the Portuguese CdLS patient cohort…………………………………………………………….44

Figure 5.2: Flowchart for a proposed procedure for the genetic molecular diagnostic of CdLS……………………………………………………………………………………………46



Abreviations:

cDNA - Complementary DNA

CdLS - Cornelia de Lange Syndrome

dATP - Deoxyadenosine triphosphate

cCTP - Deoxycytidine triphosphate

dGTP - Deoxyguanosine triphosphate

dTTP - Deoxythymidine triphosphate

dNTP - Deoxyribonucleotide

DNA - Deoxyribonucleic acid

dbSNP - Single Nucleotide Polymorphism Database

gDNA - genomic DNA

hrMCA - High Resolution Melting Curve Analysis

MLPA - Multiplex Ligand Probe Amplification

RNA - Ribonucleic Acid

LSDB - Locus Specific Database

NCBI - National Center for Biotechnology Information



1

1. Introduction

1.1. Cornelia de Lange Syndrome

Cornelia de Lange Syndrome (CdLS) is a rare, multisystemic developmental

disorder that can be characterized by a typical phenotype, although it can be highly

variable between affected individuals, depending on the severity (Krantz et al., 2004). It

has an estimated prevalence of about 1:10 000 individuals (Kline et al., 2007).

However, since additional patients with milder CdLS phenotypes have been reported in

the literature its actual incidence and prevalence may be more common than initially

suspected.

1.1.1. History and Nomenclature

The first known description of CdLS was made by Vrolik in 1849, who reported a

case as an extreme example of oligodactily (Oostra et al., 1994). In 1916, a German

physician named Winfried Brachmann made a detailed description of a case of a

patient that had died from pneumonia at 19 days of age, which presented CdLS

features (symmetric monodactyly, antecubital webbing, dwarfism, cervical ribs, and

hirsutism) (Deardorff et al., 2011; Johns et al., 2012). Seventeen years later, Cornelia

Catharina de Lange, a Dutch pediatrician described two unrelated patients that shared

similar features. Both cases had mental retardation, microcephaly and unusual facies

that resembled each other. She described these cases in detail and proposed a

diagnostic criteria, reporting these cases as typus degenerativus amstelodamensis

(Amsterdam type degeneration), after the city that she worked in (de Lange 1933;

Kumar et al., 1985).

This syndrome is usually known as Cornelia de Lange Syndrome, honoring the

pediatrician’s contribution to the characterization of the disorder, but can be also known

as Brachmann de Lange syndrome, Amsterdam type degeneration or Amsterdam

dwarfism (Johns et al., 2012).



2

1.1.2. Clinical features

CdLS phenotype includes distinctive facial dysmorphic features, hirsutism, growth

and psychomotor developmental delay, limb abnormalities, and relatively frequent

gastrointestinal and congenital heart defects (Krantz et al., 2004; Tonkin et al., 2004).

This syndrome is usually sporadic, but there are some familial occurrences reported in

about 1% of cases (Lalatta et al., 2007).

The characteristic dysmorphic facial features, are the most consistent and

recognizable clinical findings in CdLS (Liu and Krantz, 2009). This include arched

eyebrows with synophrys (and sometimes a severe ptosis), short neck, hirsute

forehead, long and thick eyelashes, and a long and smooth philtrum (Kline et al., 2007).

Ears may be lowset and posteriorly rotated, and the mid face flattened with depressed

nasal bridge. The patients can also have upper thin lips, high palate, broadly spaced

teeth and micrognatia (Liu and Krantz, 2009).

Limb defects help to corroborate a CdLS diagnosis (Kline et al., 2007). These

findings can range from small hands and feet, to severe anomalies of the upper limbs.

The majority of CdLS cases can have brachydactyly and clinodactyly, a shorter first

metacarpal, and proximally placed thumbs, while the most severe cases can have

oligodactyly, ulnar deficiencies or absent forearms (Kline et al., 2007; Krantz et al.,

2004).

Growth and developmental delay is usually observed, with patients having small

stature, microcephaly, mental and/or learning disabilities, especially speech and

language problems (Bork et al., 2004; Liu and Krantz, 2009).

Multiple organ systems can be involved in CdLS, but the mostly affected include

the gastrointestinal system (especially the presence of gastroesophageal reflux

disease), hearing, genitourinary system, ocular system, heart and diaphragm. Besides

limb defects other orthopedic complications may occur (Kline et al., 2007; Dorsettt and

Krantz, 2009).

Most patients have behavioral problems that can be caused or enhanced by the

disease physical complications. Common behavioral and neurological problems that

have been reported in CdLS patients include autistic behavior, self injurious tendencies,

obsessive compulsive behavior, attention deficit disorder, and depression (Liu and

Krantz, 2009).



3

The variability of the CdLS phenotype expression been shown since the earliest

reports of this syndrome. While the most severe cases of CdLS have many of the

symptoms described above, in the mildest forms the patients usually have mild to

moderate mental disability and (at least at a young age) typical facial characteristics,

without other substantial structural differences. This implies that the brain is the most

sensitive organ to this condition (Dorsettt and Krantz, 2009). This variation between

affected individuals has led to the development of scoring systems that help to

establish a genotype-phenotype correlation, and to anticipate care measures that will

be needed by each patient (Kline et al., 2007).

1.1.3. Etiology

CdLS is a genetically heterogeneous syndrome since five genes have been

reported as implicated in this disease. The majority of patients have de novo dominant

mutations in NIPBL gene (40-60% of patients) (Russo et al., 2012). This explains the

reason why the bulk of patients have a sporadic inheritance pattern and there are only

a few familial cases. In addition, a limited number of patients have been described as

having mutations in other genes such as SMC1A and SMC3 (~5% and <1% of patients

respectively). All of these encode for proteins involved in the Cohesin sister chromatid

cohesion (Galendari et al., 2011).

Very recently, another gene involved in the cohesion complex, HDAC8, has been

reported as causing CdLS but mutations were identified in a very small percentage of

patients (Deardorff et al., 2012a). A closely related gene, RAD21, has also been found

to cause a cohesinopathy with several overlapping features with CdLS (Deardorff et al.,

2012b). Other unidentified genes are thought to be involved with this syndrome since

many patients are still molecularly unresolved (Russo et al., 2012). Considering the

importance of the cohesin complex to this syndrome, this subject will be covered in

more detailed in the next section of this Introduction.

1.2. Biochemistry basis of CdLS - the cohesin complex

Cohesin complex has a crucial role in cell division by regulating sister chromatid

cohesion during mitosis and meiosis. During the last years several studies have



4

demonstrated that this complex, together with their key regulators, has other essential

functions during cell cycle progression. These include: i) DNA double strand repair in

G2 phase, ii) gene expression by determining the amount of elongating RNA

polymerases on genes (Fay et al., 2011) and thereby, iii) contributing for maintaining

the genome’s stability (reviewed by Horsfield et al., 2012).

During mitosis and meiosis cohesin complex is composed by several proteins

(Figure 1.1 and Table 1.1). It comprises a heterodimer of SMC1A and SMC3, forming a

“hinged” domain. This dimer interacts via the ATP-binding regions with α-kleisen

subunit (RAD21), which subsequently interacts with Stromalin/SA subunit (known

STAG), forming a ring-like structure (Hosfield et al., 2012). The prevailing theoretical

model suggests that these molecular “rings” entrap sister chromatics by encircling DNA.

A new layer of complexity is added to this model when considering that the cohesin’s

“life-cycle” (turnover, recycling, loading and unloading onto chromosomes) is regulated

by several molecular partners (Table 1.1). The loading of cohesin to DNA is controlled

by NIPBL/MAU2 complex, whereas the opposite activity is mediated by a “releasin”

(WAPAL/PDS5) which interacts with SA to open the ring structure. In humans, ESCO1

and ESCO2, two cohesion acetyl transferases, are necessary for SMC3 acetylation

(Hosfield et al., 2012). This step is necessary for adequate sister chromatid cohesion,

since generates the “cohesive” form of cohesin. Interestingly, defects in ESCO2 are

linked to another cohesinopathy, known as Roberts Syndrome (RBS, MIM#268300)

(Vega et al., 2005). SMCs complexes can also be “recycled” and reloaded to chromatin

after mitosis. An essential step for this recycling is the deacetylation of SMC3 by

HDAC8, which is the opposite function of ESCO2.

Figure 1.1 – Schematic representat ion of the cohesin complex during mitosis,

showing the different core components and regulators (adapted from Horsfield et al.,

2012).



5

Table 1.1- Members and regulators of the cohesin complex (proteins, genes and phenotypes ).

#? - omim number is not available yet; CDLS- Cornelia de Lange Syndrome.

Cohesin complex

Protein Main function Gene, Locus, MIM*

Phenotype, MIM#

Key references

SMC subunits

Structural maintenance of chromosomes 1A (SMC1A)

Core cohesin subunit

SMC1A , Xp11.22-p11.21, *300040

CDLS2, #300590

Sumara et al., 2000; Musio et al., 2006

Structural maintenance of chromosomes 1B (SMC1B)

Cohesin subunit (meiosis)

SMC1B , 22q13.31, *608685

- Revenkova et al., 2001; Revenkova et al., 2004

Structural maintenance of chromosomes 3 (SMC3/CSPG6/ Bamacan)


SMC3 , 10q25, *606062

CDLS3, #610759 Sumara et al., 2000;

Deardorff et al., 2007

α-Kleisin subunit

Homolog of RAD21 S. pombe (RAD21)


RAD21, 8q24, *606462

CDLS4, #614701 Sadano et al., 2000;

Deardorff et al., 2012b

Homolog of REC8 S. pombe (REC8)


REC8, 14q11.2-q12, *608193

- Parisi et al., 1999

Stromalin/ SA subunit

Stromalin antigen 1 (STAG1/SA1/SCC3A)

Cohesin subunit

STAG1 , 3q22.3, *604358

- Carramolino et al., 1997; Sumara et al., 2000

Stromalin antigen 2 (STAG2/SA2/SCC3B)

STAG2, Xq25, *300826

- Carramolino et al., 1997 Sumara et al., 2000

Stromalin 3 (STAG3/SA3)


STAG3, 7q22.1, *608489

- Pezzi et al., 2000

Interactors of α-kleisin and SA

Homolog of PDS5A, regulator of cohesion maintenance A S. cerevisiae ( PDS5A)

Balancing cohesion establishment with cohesin dissociation

PDS5A, 4p14, *613200

-

Kumar et al. 2004 ; Zhang et al., 2009

Homolog of PDS5B, regulator of cohesion maintenance B S. cerevisiae (PDS5B/APRIN/AS3)

PDS5B, 13q12.3, *605333

-

Geck et al., 2000; Zhang et al., 2009

Cell division cycle-associated protein 5 (CDCA5/SORORIN)

CDCA5, 11q12.1, *609374

- Rankin et al., 2005; Nishiyama et al., 2010

Homolog of wings apart-like protein Drosophila (WAPAL/WAPL)

WAPAL, 10q23.2, *610754

- Kueng et al., 2006; Nishiyama et al., 2010;

Kollerin Nipped-B-like (NIPBL/SCC2/ DELANGIN)

Cohesin loading

NIPBL, 5p13.2, *608667

CDLS1, #122470

Tonkin et al., 2004; Krantz et al., 2004

Homolog of MAU2 chromatid cohesion factor C. elegans (MAU2/SCC4)

MAU2, 19p13.11, *614560

-

Watrin et al., 2006

Cohesin acetyl transferase

Homolog of establishment of cohesion 1 S. cerevisiae (ESCO1)

Establishment of cohesion

ESCO1, 18q11.2, *609674

-

Hou and Zou, 2005; Zhang et al., 2008

Homolog of establishment of cohesion 2 S. cerevisiae (ESCO2)

ESCO2, 8p21.1, *609353

Roberts syndrome, #268300 SC phocomelia syndrome, #269000

Hou and Zou, 2005; Vega et al., 2005; Gordillo et al., 2008

Cohesin deacetylase

Histone deacetylase 8 (HDAC8)

Recycling of cohesin

HDAC8 , Xq13, *300269

CDLS #?

Buggy et al., 2000; van den Wyngaert et al., 2000; Deardorff et al., 2012a



6

1.3. Genetics of CdLS

1.3.1. NIPBL gene

It was in 2004 that NIPBL (MIM*608667), the first gene responsible for CdLS was

found by two independent groups (Krantz et al., 2004; Tonkin et al., 2004). This gene

was found through genome-wide linkage exclusion analysis in affected families, in

patients with de novo translocations and deletions, that allowed the identification of a

candidate genomic region for CdLS. Using distinct approaches both works identified

mutations in a gene named as NIPBL (Nipped-B like, MIM*608667) as causative of the

disease (Krantz et al., 2004; Tonkin et al., 2004).

The human NIPBL gene is located in chromosome 5, (5p13.1) and spans over

190Kbp. The open reading frame (ORF) of NIPBL starts in exon 2, and continues to

exon 47 (Tonkin et al., 2004). This gene encodes for delangin, a protein which contains

a highly conserved armadillo domain with several HEAT-repeats, a protein-protein

interaction motif (Selicorni et al., 2008), a glutamine-rich domain, and a predicted

bipartite nuclear localization signal (Yan et al., 2006; Braunholz et al., 2012). This

protein has two isoforms: delangin-A (the mainly expressed form), a long form with

2804 amino acids, and delangin-B, the shorter form with 2697 amino acids (Tonkin et

al., 2004). It is involved in cohesin complex loading, mediating chromatin modifications

through recruitment of histone deacetylases. NIPBL was identified as the human

homologue of Drosophila melanogaster Nipped-B gene, a chromosomal adherin

involved in chromatid cohesion and also acting as a transcriptional regulator in Yeast

(SCC2) (Bhuiyan et al., 2006). Further studies have demonstrated that the human

protein is a regulator of the cohesin complex required for binding of the cohesin

complex (Braunholz et al., 2012; Dorsett et al., 2007). NIPBL-mutated cells from CdLS

patients have shown a reduced capacity to tolerate DNA damage (Revenkova et al.,

2009).

Nearly half of CdLS cases are caused by mutations in NIPBL gene, being the

major causative gene for this syndrome. Mutations in this gene can cause mild to

severe forms of the disease (Galendari et al., 2011). Genotype-phenotype correlations

studies indicate that severe NIPBL mutations (originating premature termination

codons) usually cause more severe forms than missense mutations (Dorsettt and

Krantz, 2009).



7

0

2

4

6

8

10

12

14

16

18

20

5'U

TR-1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46

47-3

'UTR

Distribution of mutations NIPBL gene

Total mutations (count) Mutations [(count/exon size)*100]

56

In 2010, Oliveira et al., reported the creation of a locus specific database (LSDB)

for the NIPBL gene, which contained mutational and clinical data reported in all peer

review papers. An analysis of the mutations in the database was made, and all

mutations types were found in NIPBL: small deletions (27.6% of cases), missense

changes (21.1%), nonsense mutations (17.3%), mutations affecting splicing (17.1%),

small duplications (13.6%), insertions (1.5%) and insertion/deletions (1.5%). Statistical

analysis suggested that patients with nonsense mutations or out-of-frame

deletions/duplications/insertions more often presented a severe phenotype than those

with less detrimental effect on the protein (Oliveira et al., 2010).

The mutations in NIPBL seem to be spread all over the gene, without a clear

hotspot. However, the authors have shown that mutation frequencies vary throughout

the gene, meaning that some gene regions seem to be more prone to mutations than

others (Figure 1.2). Gross genomic duplications or deletions in NIPBL gene are

considered rare, and there were only a limited number of studies concerning their

occurrence (Bhuiyan et al., 2007; Russo et al., 2012).

Figure 1.2 – NIPBL-LOVD database content analysis, showing the distribut ion of mutat ions and

their frequency. Adapted from Oliveira et al. (2010).

1.3.2. SMC1A gene

After the identification of NIPBL as a causative gene for CdLS, it was realized that

mutations in this gene are only present in about half the patients. This mutation

detection rate suggested that there could be other genes involved with this syndrome.



8

The variability of the clinical findings between patients also supported this hypothesis.

Since NIPBL is involved in the cohesion complex, other genes encoding proteins linked

to the cohesion complex were studied in patients with CdLS or CdLS-like phenotype.

In 2006, mutations in the SMC1A gene (structural maintenance of chromosomes

1A, MIM #300590) was found to cause an X-linked form of CdLS patients (Musio et al.,

2006). Mutations were found in three affected males of the same family and in an

unrelated patient, without NIPBL mutations (Musio et al., 2006). The SMC1A gene is

located in chromosome Xp11.22, with a size of 48Kbp, and has 25 exons (all of them

coding). The SMC1A gene encodes for a protein with 1233 aminoacids, which is one of

the four subunits of the cohesion complex. It is involved in genome stability and DNA

repair (Musio et al., 2006; Watrin and Peters, 2006). Interestingly, the majority of

patients with SMC1A mutations are females (Deardorff et al., 2007). While in normal

females the SMC1A gene escapes X-inactivation and both copies are expressed,

males are hemizygous for SMC1A since it is not located in the pseudoautosomal

region. Affected females should express a normal copy of the gene, leading to the

suggestion that SMC1A mutations are dominant-negative (Liu and Krantz, 2009;

Hoppman-Chaney et al., 2011).

X-linked CdLS patients seem to have a milder phenotype than the “classic” form,

especially the female patients, having a milder dysmorphism, and a mild or no growth

delay, and absence of microcephaly (Musio et al., 2006; Bork et al., 2007). Less severe

cases result in a phenotype that approaches that of apparently nonsyndromic mental

retardation (Revenkova et al., 2009), although severe phenotypes in this gene have

also been reported (Hoppman-Chaney et al., 2011). Male patients seem to have a

severer phenotype than females, more similar with a “typical” CdLS. (Musio et al.,

2006; Maninni et al., 2010).

A LSDB for SMC1A gene was also developed (Oliveira et al., 2010). Most

mutations reported were substitutions (~84%), and the remaining are small deletions

(~16%). The majority of the mutations reported in this database have a de novo origin.

The mutations appear spread over the gene, without a clear mutational hotspot

(http://www.lovd.nl/SMC1A, last accessed 15 September 2012).



9

1.3.3. SMC3 gene

The discovery that mutations in the X-linked SMC1A gene result in CdLS

suggested that other members of the cohesin complex may contribute to the etiology of

CdLS. In 2007, Deardorff and collaborators screened 115 CdLS patients that were

negative for mutations in NIPBL, for mutations in genes encoding for subunits of the

cohesin complex (SMC1A and SMC3). There was only one mutation found in SMC3

gene so far, in two unrelated patients. It is a small deletion in exon 9 that results in the

loss of a single amino acid. (Deardorff et al., 2007; Revenkova et al., 2009).

The SMC3 gene (MIM#610759), or structural maintenance of chromosomes 3, is

located on chromosome 10q25.2, and it has a size of 37Kbp, and 29 exons (all of them

coding). This gene encodes for the SMC3 protein, with 1217 aminoacids, and like

SMC1A, is one of the subunits of the cohesion complex (Deardorff et al., 2007;

Revenkova et al., 2009). Patients with mutations in SMC3 present a mild variant of the

CdLS phenotype, with very mild facial features, and no absence or reduction of limbs

or digits, and no other major structural anomalies in contrast to classical CdLS

phenotype (Deardorff et al., 2007).

1.3.4. HDAC8 gene

The involvement of HDAC8 in CdLS was discovered very recently by Deardorff and

co-workers (2012a). His group characterized this gene product as a SMC3 deacetylase,

and also detected six loss-of-function mutations in HDAC8 (Deardorff et al., 2012a).

The HDAC8 gene (MIM#601639), or histone deacetylase 8, is located in

chromosome Xq13.1, has a size of 243.6 Kbp, and contains a total of 61 exons. In the

cohesion complex, the HDAC8 protein is the SMC3 deacetylase. The SMC3 protein is

acetylated by ESCO2 during S-phase to establish cohesiveness of chromatin-loaded

cohesion, and HDAC8 protein deacetylates SMC3 during anaphase. The loss of

HDAC8 activity in increased SMC3 acetylation and inefficient dissolution of the ‘used’

cohesin complex released from chromatin in both prophase and anaphase. Deardorff

and team suggested that this defects lead to impaired embryonic development which

can give rise to CdLS. (Deardorff et al., 2012a).



10

2. Objectives

The molecular and clinical characterization of CdLS patients was initiated at the

national level in 2005 with the research project entitled: “Caracterização clínica,

epidemiológica e molecular dos doentes com o Síndrome de Cornelia de Lange em

Portugal”, supported by the Instituto de Genética Médica Dr. Jacinto Magalhães. Our

team of molecular and clinical geneticists gathered mutational and clinical data on a

total of 53 Portuguese CdLS patients (this includes more recent cases). Thirteen of

those patients, with novel NIPBL mutations, were initially reported together with the

development of a LSDB (Oliveira et al., 2010). The present work intended to unfold

new molecular findings in the remaining Portuguese CdLS patients. In brief, the main

objectives of the present work were the following:

i) Report the molecular characterization of the remaining CdLS patients with

NIPBL mutations.

ii) Apply Multiplex Ligation-dependent Probe Amplification technique (for

NIPBL) in the group of CdLS patients that remained molecularly unresolved

after NIPBL gene sequencing.

iii) Set up a fast and cost-effective screening method for point mutations in

SMC1A gene by High resolution melting curve analysis.

iv) Optimize the molecular diagnostic procedure of CdLS patients.



11

3. Materials and Methods

3.1. Patients

Forty CdLS Patients were included in this work, recruited in two distinct phases:

i) Research project (2005-2010).

The criteria for inclusion were a diagnosis of CdLS confirmed by a clinical

geneticist, a normal karyotype and guardian informed consent. A clinical protocol

designed for a previous study (Dias et al., 2008) was used by clinical geneticists to

assess the patients. This protocol includes information on prenatal, birth and postnatal

history, an evaluation of 50 phenotypical features, behavior, development and

multisystem involvement. A diagnostic severity score was also included in the protocol

formulary. This scoring system described by Kline and collaborators (2007), classifies

CdLS patients as mild, moderate or severe, according to the number and type of

abnormalities (Annex I). From the 53 patients (from 51 different families) initially

referred, 26 patients were recruited for this work. Fourteen were excluded for one of

the following criteria: insufficient DNA for analysis (deceased patients, n=2), insufficient

clinical data (incomplete/lacking clinical protocol information, n=2), patients with low

severity score (with “Cornelia-like” features but not reaching the mild phenotype, n=10).

Thirteen of these patients (from 12 different families) were previously published by our

group (Oliveira et al., 2010) and thus, will not be included in the results section of this

work.

ii) Clinical requests for CdLS molecular studies (2010-2012).

Fourteen new CdLS cases were referred for molecular studies. Together with the

biological samples a specific request form was used to collect as most as possible

information concerning their phenotype. The clinical classification of patients was

based on the information retrieved from this test request form and was validated by a

clinical geneticist using the score system proposed by Kline et al. (2007).



12

3.2. Preparation of biological samples

3.2.1. DNA extraction method

Genomic DNA (gDNA) was extracted from peripheral blood samples, collected

from the patients and available parents in EDTA anticoagulant tubes, using salting out

method (Miller et al., 1988). Since this method does not requires the use of toxic

chemicals it is a good alternative to the phenol/chloroform extraction. Salting-out is

based on the principle that proteins and other cellular contaminants are less soluble at

higher salt concentrations due to their relative hidrophobicity and will precipitate in a

saturated salt solution while the DNA does not.

3.2.2. RNA extraction method

One of the major concerns in manipulating RNAs, are the ribonuclease enzymes

present in cells and tissues that can rapidly destroy these nucleic acids. Thus, it is

important to avoid sample contamination with ribonucleases. Nuclease-free pipette tips

and reagents were used and all micropipettes (used exclusive for RNA extraction),

were cleaned with RNase AWAY™ (Life Technologies, Foster City, CA, USA). Total

mRNA purification was made from peripheral blood sample in EDTA anticoagulant tube,

using the commercial kit PerfectPure™ RNA Purification System (5 PRIME, Hamburg,

Germany) following the manufacturer protocol.

3.2.3. Nucleic acid quantification

After the nucleic acid extraction, purity and concentration were determined through

spectrophotometric absorbances at the wavelengths of 260 nm (A260) and 280 nm

(A280). A 2 µl sample was quantified in a Nanodrop spectophtometer (Thermo

Scientific), with two independent measures. A260 is used to measure DNA (1U A260 =

50 ng/μl) and RNA (1U A260 = 40 ng/μl) concentrations, while A280 is used to

measure protein concentration. The A260/A280 ratio is used to determine the purity of

nucleic acids. A pure DNA and RNA sample would have a ratio of ~1.8 and ~2.0

respectively, while a protein contaminated sample has a lower ratio (Maniatis et al.,

1982) (Glasel, 1995).

http://en.wikipedia.org/wiki/Ribonuclease



13

3.2.4. RNA conversion to cDNA

Since RNA is so easily degraded, it was more convenient to convert it to cDNA

(complementary DNA). The High-Capacity cDNA Archive Kit (Life Technologies, Foster

City, CA, USA) was used for the conversion, following the manufacturer protocol. This

was made through a reverse transcription reaction (RT reaction), using MultiScribe™

Reverse Transcriptase (Life Technologies, Foster City, CA, USA) enzyme, and random

(examer) primers. The RNA template was mixed with the kit reagent mixture, and

placed in a thermocycler, and placed 10 minutes at 25ºC and 120 minutes at 37ºC.

3.2.5. Aliquots preparation

The gDNA samples were aliquoted at 100 μg/ml and 50 μg/ml concentrations for

PCR (section 3.3), and for high resolution melting curve analysis (hrMCA) (section 3.8)

use, respectively. This step was essentially done to preserve the original stock sample

for future analysis, and to set the same gDNA concentration between all samples. This

is not only useful to optimize the PCR but also critical for MLPA (section 3.6) and

hrMCA techniques by prevent bias and allowing accurate results. The aliquots and

original samples were stored at 4ºC until use, thus avoiding freeze-thaw of the samples

at lower temperatures).

3.3. DNA amplification by PCR

The polymerase chain reaction (PCR) is an in vitro technique used in molecular

biology, to amplify a target sequence (amplicon) of DNA exponentially. Besides the

DNA template being amplified, the PCR mixture needs several components, including:

Oligonucleotide sequences (primer pair) with a sequence complementary to

the 3' (3 prime) ends of each of the sense and anti-sense strand of the DNA

target region.

Deoxynucleotide triphosphates - dATP, dCTP, dGTP and dTTP (dNTPs).

DNA polymerase, (such as the thermostable Taq polymerase).

http://en.wikipedia.org/wiki/Directionality_%28molecular_biology%29

http://en.wikipedia.org/wiki/DNA#Sense_and_antisense

http://en.wikipedia.org/wiki/DNA_polymerase



14

Buffer solution that enables the suitable activity and stability of the DNA

polymerase.

Divalent cations (like Mg2+) and monovalent cation (like K+).

This technique consists of cycles of: denaturation (separation of double strand

DNA into single stranded DNA), annealing (primers pair with DNA complementary

sequence), and extension (enzymatic replication of the DNA). This is done by

repeatedly heating and cooling the reaction mixture. The DNA polymerase uses a

single DNA strand to create a new complementary DNA strand in the site targeted by

the pair of primers. The new DNA is itself used as a template for replication, creating a

chain reaction in which the DNA template is exponentially amplified. Besides the cycle

phase, there are some PCR programs that use a preliminary first denaturing step,

heating the sample for heat-activated enzymes, or to help denature the target DNA

better. A final extension (or final elongation) step can also be used to make sure that

the single-stranded DNA is entirely extended (Maniatis et al., 1982).

3.3.1. Oligonucleotides design

Primer design is a critical step for a successful PCR approach. When designing

primers several parameters should be considered (Burpo, 2001):

The primer sequences should be unique to target only a region of DNA, and

avoid annealing at a similar sequence.

Primers with long repeats of a single nucleotide should be avoided (loop

formations can occur).

Primers should not anneal with other primers or themselves (formation of

primer dimers could contaminate the sample).

Primers should be located at least 30-40 bases upstream of the region of

interest in the sequence read.

Optimal primer GC content between 40-60%

NIPBL gene

The mainly expressed isoform of the NIPBL gene (transcript with the accession

number NM_133433.3) has a total of 47 exons, 46 of which are coding (exon 1 is non-

coding). A total of fifty-four primer pairs were used to completely cover the coding and

flanking regions of the gene. In addition to the primers described by Krantz et al. (2004),

http://en.wikipedia.org/wiki/Enzyme

http://en.wikipedia.org/wiki/DNA_replication

http://en.wikipedia.org/wiki/Chain_reaction

http://en.wikipedia.org/wiki/Exponential_growth



15

used for exon 10 and exons 39 to 47 of NIPBL gene, custom primers were designed for

the remaining gene regions (exons 2 to 9 and 11 to 38, Annex II.a).

SMC1A gene

According to the transcript with accession number NM_006306.2, SMC1A gene

has a total of 25 exons. Twenty six primer pairs were designed and used to cover the

coding and flanking intronic regions of the gene (Annex II.b). The primers designed

were optimized for hrMCA technique, with smaller amplicons, and without M13

universal tail.

Primer Express (Life Technologies, Foster City, CA, USA) software was used for

primer design. Besides speeding up this process the application allowed to

standardization of experimental conditions. Primers were chosen to have the same

annealing temperature (58-60ºC). In the case of NIPBL gene approach primers

incorporated a M13 universal primers tail sequence. Since the M13-tailed sequences

are not present in the Human genome, primers are not fully complementary (at their 5’).

These M13 universal primers are useful for downstream applications (such as Sanger

sequencing) since the same sequencing primer can be used for all amplicons. In order

to exclude primers more susceptible to primer-dimer formation, Fast PCR software

(Kalendar and Schulman, 2009) was used. Finally, primer-BLAST (Ye et al., 2012,

accessed at http://www.ncbi.nlm.nih.gov/tools/primer-blast/) algorithm was used to

check for primer specificity.

3.3.2. PCR mixture and thermocycling profiles

The NIPBL PCR mixture (for all primer pairs) was prepared with the following

reagents:

10 μl of PCR Master Mix 2x (Cat No # M7505, Promega, Fitchburg, WI, USA)

7 μl dH2O (nuclease free)

1 μl Forward Primer + 1 μl Reverse Primer (primers at 10 pmol/μl)

1 μl gDNA (100 ng/l)



16

Then, the PCR mixture samples were homogenized and placed in a thermocycler

(Veriti® Thermal Cycler 96-well, Life Technologies, Foster City, CA, USA), to run one of

the NIPBL programs.

While all the other exons use the same PCR program (Table 3.1) with only

differences in annealing temperatures, exon 44 primer pair had to be performed using

a different thermocycler profile. This is mainly due to the particular base pair

composition of this amplicon, which is particular prone to nonspecific amplification and

a high background. Also, it was difficult to find a new alternative primer pair. To

overcome this problem, a Touchdown PCR program (Table 3.2) was used for exon 44.

This particular type of PCR increases the specificity of primer annealing by initiating at

with higher annealing temperatures and by gradually reducing it. Higher temperatures

result in greater specificity for the binding between the primer and the targeted DNA

template, but low amplification yield. As cycle-by-cycle the annealing temperature

lowers resulting in less specificity but increasing the efficiency of amplification. Due to

the exponential amplification nature of PCR, the amplified (specific) sequences during

the first cycles will out-compete the nonspecific sequences that may be result at lower

temperatures.

Table 3.1: PCR program for NIPBL ampli f icat ion (except for exon 44)

Step Temperature (ºC) Time Number of cycles

initialization 94 5 min 1

denaturing

annealing

expansion

94

Ta

72

30 sec

30 sec

45 sec

38

final extension 72 10 min 1

final hold 15 - -

Ta - annealing temperature: for the custom primers and primers 10A, 40, 42, and

45 to 47B, the annealing temperature was 58ºC. For the primers 10B to 10F and 39 the annealing temperature was 55ºC, for the primer pair 43, the annealing temperature was 52ºC.



17

Table 3.2 : Touchdown PCR program for NIPBL exon 44

Step Temperature (ºC) Time Number of cycles

initialization 94 5 min 1

denaturing

annealing

expansion

94

62 → 54 *

72

30 sec

30 sec

45 sec

10

denaturing

annealing

expansion

94

54

72

30 sec

30 sec

45 sec

25

final extension 72 10 min 1

final hold 15 - -

* -0.8ºC each cycle, starting on the second cycle

3.4. Agarose gel electrophoresis

After PCR amplification PCR products were analyzed by agarose gel

electrophoresis. Gel electrophoresis is a way to separate and analyze DNA. It can be

used to quantify or even to isolate a particular DNA fragment. An electric field causes

negatively charged DNA molecules to travel through the agarose gel. DNA molecules

must move through the matrix pores of the gel allowing their efficient separation

according to size (being easier for the smaller molecules to migrate than the larger

ones). To verify the PCR amplification 3 μl of each PCR sample product were mixed

with 1 μl loading buffer (custom made) and loaded in 1% (m/v) SeaKem® LE Agarose

(Cambrex Bio Science, Verviers, Belgium) gel stained with GelRed™ (Biotium,

Hayward, CA, USA). A DNA low mass ladder (Life Technologies, Foster City, CA,

USA) was also loaded in a separate lane of the gel to calculate the size to the DNA

fragments. After running electrophoresis the gel was analyzed under UV light and

images were obtained using a LAS-3000 image system (Fujifilm, Tokyo, Japan).



18

3.5. PCR purification

The PCR product was purified using the ExoSAP-IT® (Affimetrix, Santa Clara, CA,

USA) enzymatic treatment. ExoSAP-IT® is a mixture of shrimp alkaline phosphatase

and exonuclease that degrade primers and dephosphorylate dNTPs, improving the

quality of the sequences resulting from PCR product.

In this procedure 2 μl of ExoSAP-IT® was added to 8 μl of PCR product (for each

amplicon) and this mixture was placed for 30 minutes at 37ºC followed by 15 minutes

at 85ºC (enzyme inactivation step).

3.6. Sanger Sequencing

Purified PCR products were sequenced by dye-terminator cycle sequencing. In this

type of sequencing, there are 4 (one for each base) chain terminator ddNTPs

(dideoxynucleotide-tri-phosphate) tagged with fluorescent dyes that emit light in

different wavelengths. These ddNTPs are mixed in lower concentration with dNTPs

and when a DNA polymerase incorporates a ddNTP instead of a dNTP the chain

extension is prematurely ended. This will produce a series of DNA fragments with

different lengths that are terminated with a specific base. Cycle sequencing products

must be cleaned of primers, excess dNTP’s, enzymes and buffer components before

proceeding for electrophoresis. Currently this analysis is carried out in multi-capillary

automated sequencers that can efficiently analyze several samples at the same time.

Briefly, samples are injected into the capillaries loaded with a special polymer (used as

a separation matrix) by means of an electric potential difference. This capillary

electrophoresis resolves each dye labeled DNA fragments with high efficiency (1 bp

difference). At the end of the capillary a laser-induced fluorescent system is used to

excite each fluorochrome and the resulting light is detected. Finally the software

converts all the fluorescence signals into an electropherogram.

3.6.1. Sequencing reaction

This sequencing reaction is an asymmetric PCR carried out using only one primer,

resulting in the amplification of only one DNA strand of the double-stranded DNA



19

template. This means that to sequence both DNA strands two independent sequencing

reactions are needed for each amplicon (one for each of the primer pair).

Sequencing reactions were prepared with BigDye® Terminator Cycle Sequencing

Kit v1.1 (Life Technologies, Foster City, CA, USA), using a mixture of 2 μl big-dye, 2 μl

dH2O, 1 μl primer (at 5 pmol/μl) and 5 μl purified PCR products. The M13 Universal

primers (Forward and Reverse) were used in the sequencing reaction (instead of the

PCR primers) for the M13-tailed primers. For amplicons without M13-tailed primers, the

initial PCR primers were used.

3.6.2. Sequencing PCR purification

After cycle sequencing, the samples were purified by Performa® DTR Ultra 96-

Well Plates (Edge Bio, Gaithersburg, MA, USA). The samples were transfered into the

Performa® well plates, that were centrifuged at 850 rcf in an Eppendorf 5804 R Plate

Centrifuge (Eppendorf, Hamburg, Germany). These columns have a hydrated gel

matrix that purified the samples of dye terminators, dNTPs, primers, and buffer.

3.6.3. Sequencing analysis

The fragments were separated by size through capillarity electrophoresis in 3130xl

ABI Genetic Analyser (Life Technologies, Foster City, CA, USA) and sequence

analysis was performed using SeqScape® v2.5 software (Life Technologies, Foster

City, CA, USA). Identified sequence variants were confirmed in a secondary

confirmatory PCR of the affected amplicon, in the patient and parents (if available).

3.7. Multiplex ligation probe amplification

The multiplex ligation-dependent probe amplification (MLPA) is a semi-quantitative

technique that uses a primer pair to amplify several target DNA locations, by using

probes. This technique allows the detection of gross deletions and duplications in the

target DNA sequence. Each of the probes is composed of two oligonucleotides that

target adjacent sites. One of them has a tail sequence complementary to the forward



20

primer, and the other to the reverse primer. Only when both oligonucleotides are

hybridized with their respective target sequence, they can be ligated into a complete

probe. Each of the complete probes has a unique size, which can be identified by

capillarity electrophoresis. By using reference samples it is possible to compare the

relative quantity of each of the fragments.

The MLPA analysis was performed using commercial kits P141 and P142 (MRC-

Holland, Amsterdam, the Netherlands), that contain probes for all NIPBL exons. First, a

mixture was prepared with the DNA samples and the NIPBL probes, and subsequently

heated to denature followed by overnight hybridization. After this period a DNA ligase is

added to the mixture to uphold probes’ ligation. These ligation products are finally

amplified by PCR. Final MLPA products were separated by capillary electrophoresis,

using a 3130xl ABI Genetic Analyser (Life Technologies, Foster City, CA, USA).

Results were analyzed using GeneMarker® software (SoftGenetics LLC, State College,

PA, USA). The population normalization method was used and data was plotted using

probe ratio.

3.8. High Resolution Melting Curve Analysis

The hrMCA is a technique used to detect sequence variants in double stranded

DNA samples. During gDNA amplification by PCR, an intercalating fluorescent dye is

added at a saturating concentration, and its incorporation and binding to double

stranded DNA causes it to become florescent. The changes in fluorescence are

detected by an optical system during real-time. After the PCR step, the specific hrMCA

thermocycling profile is initiated, samples are heated from between two temperature

(such as from 50 to 95ºC) in a precise and uniform way. When the specific melting

temperature of the amplicon is reached, the two DNA strands separate (melt) from

each other, releasing the intercalated dye which loses its fluorescence. Since the

melting temperature of an amplicon depends on its DNA composition, a single base

pair change in the sequence will shift the temperature at which the melting occurs. By

comparing melting curve profiles of wild type controls and from patients it is possible to

identify sequence variants. This technique was used to detect variants in SMC1A gene,

and for population screening in some NIPBL variants.

The PCR and subsequently hrMCA were prepared with the following reagents:



21

10 μl of PCR Master Mix 2x (Promega Cat No # M7505)

6 μl dH2O (nuclease free)

1 μl EvagreenTM (Biotium, Hayward, CA, USA)

1 μl Forward Primer + 1 μl Reverse Primer (primers at 5 pmol/μl)

1 μl gDNA (50 ng/l)

The samples were placed in a Rotorgene 6000 (Corbett Life Science, Sydney,

Australia) where the hrMCA program was performed (Table 3.3). The run data were

analyzed in the Rotorgene 6000 software, and any variation from normal curve pattern

was sequenced and analyzed (using the same procedures described in 3.6.2. and

3.6.3. but to SMC1A gene).

Table 3.3 : High Resolut ion Melt ing Curve Analysis for SMC1A gene

Step Temperature (ºC) Time Number of

cycles

PCR

phase

Initial

denaturing 94 2 min 1

denaturing

annealing

expansion

94

56

72

20 sec

30 sec

30 sec

35

Final extension

72 5 min 1

Melting

phase

preparation 50 5 min

melting 50 → 95 * -

* on melting phase, temperatures rise gradually, in steps of +0.05ºC followed by a 2 second hold to stabilize and register the fluorescence level



22

3.9. Bioinformatics and databases

The variations discovered through the different techniques, were analyzed with the

aid of Alamut 3.0 (Interactive Biosoftware, Rouen, France) mutation interpretation

software, and the locus specific mutation databases (LSDB) for NIPBL and SMC1A

genes (http://www.lovd.nl/CDLS). These databases use the Leiden Open Variation

database (LOVD) software platform, enabling a Web-based listing and curation of

sequence variations associated with phenotypical information.

3.9.1. Missense mutations analysis

The bioinformatic analysis of the pathogenicity was made with the help of the

commercial software Alamut version 3.0 (Interactive Biosoftware, Rouen, France), and

three other tools:

PolyPhen version 2. This software gives a score to the impact of

substitutions on the structure and function of protein, with a range of

values from: 0 to 1.000, with the higher value more likely to be

pathogenic (Adzhubei et al., 2010).

SIFT is a sequence homology-based tool that predicts whether an

amino acid substitution in a protein will have a phenotypic effect. The

score ranges from 0 to 1. The substitution is predicted damaging is the

score is <= 0.05, and tolerated if the score is > 0.05 (Pauline and

Henikoff, 2003).

Grantham's distance compares wild type with mutated aminoacids, by

using physical and chemical parameters (volume, weight, polarity, and

carbon-composition). The score range from 0 to 215, with higher value

indicates larger difference.

3.9.2. Algorithms for evaluating splicing

The splicing mutations found were analyzed through five different methods,

integrated in the Alamut 3.0 (Interactive Biosoftware, Rouen, France) Splicing

Prediction Module:



23

SpliceSiteFinder-like method is based on position weight matrices

computed from a set of human constitutive exon/intron junctions for donor

and acceptor sites.

MaxEntScan method uses the Maximum Entropy principle (Yeo et al.,

2004).

NNSPLICE is a method based on neural networks (Reese et al., 1997).

GeneSplicer uses several techniques to detect splice sites, among which

the Markov models (Pertea et al., 2001).

Human Splicing Finder method is based on position weight matrices

with position-dependent logic (Desmet et al., 2009).

http://en.wikipedia.org/wiki/Position-specific_scoring_matrix

http://en.wikipedia.org/wiki/Position-specific_scoring_matrix



24

4. Results

4.1. NIPBL gene sequencing

Mutations in NIPBL gene were screened by direct sequencing which encompassed

all coding exons and flanking intronic sequences, the methodology is outlined (figure

4.1)

Figure 4.1: Screening for NIPBL gene mutat ions by sequencing.

A total of 11 NIPBL point mutations were detected using this approach (Table 4.1).

These variants seem to be dispersed along NIPBL gene (Figure 4.2) without a clear

mutational hotspot.



25

a) Mutations are described using NIPBL cDNA reference sequence with accession number: NM_133433.3, according with HGVS nomenclature guidelines; SEQ – Sequencing; MLPA – Multiplex ligation

probe amplification; b) CdLS clinical classification according to Kline et al., (2007); c) first report describing variant; d) somatic mosaicism suspected; e) other complete deletions of NIPBL has been

described in the literature, since the breakpoints of this mutation are still unidentified, it is not possible to known if it is the same mutation or not.

Patient Mutationa) Gene Location Predicted

polypeptide change Type

Variant/

Origin Detection

Clinical

Phenotypeb) Referencesc)

P46 c.64+1G>A Exon 2 p.(?) splicing de novo SEQ. mild Borck et al., 2004

P53 c.86del Exon 3 p.Pro29Hisfs*18 frameshift unknown SEQ. severe Novel

P31 c.1885C>Td) Exon 10 p.(Arg629*) nonsense unknown SEQ moderate Myake et al., 2005

P49 c.3316C>T Exon 12 p.(Arg1106*) nonsense de novo SEQ mild Miyake et al., 2005

P39 c.4422G>T Exon 21 p.(Arg1474Ser) missense de novo SEQ mild Oliveira et al., 2010

P34 c.5471C>T Exon 29 p.(Ser1824Leu) missense unknown SEQ mild Oliveira et al., 2010

P14 c.6653_6655delATA Exon 39 p.(Asn2218del) in-frame deletion de novo SEQ moderate Borck et al.,2004

P20 c.6763+5G>T Intron 39 p.(?) splicing unknown SEQ mild Krantz et al., 2004

P11 c.6983C>G Exon 41 p.(Thr2328Arg) missense unknown SEQ n.a. Novel

P09 c.7168G>A Exon 42 p.(Ala2390Thr) missense de novo SEQ moderate Gillis et al.,2004

P47 c.7307C>T Exon 43 p.(Ala2436Val) missense de novo SEQ n.a. Novel

P40 c.(?_-481)_(*927_?)del Exon 1 to 47 p.(?) large deletion de novo MLPA severe unknowne

P32 c.(5710-?)_(*927_?)del Exon 31 to 47 p.(?) large deletion unknown MLPA severe Novel

Table 4.1: Mutat ions found in NIPBL gene



26

5’ 3

’ exons

Patient

Figure 4.2 : Distribut ion of NIPBL mutat ions ident i f ied in this work . The blue l ines represent the int ronic areas of the gene, and the blue rectangles represent

exons. The introns are not up to scale, being much larger than represented. Numbers under the exons indicate the exon number. Green marks above the exons

represent the locat ion of the point mutat ions reported in this work, and the numbers above indicate pat ient number. Large del et ions found by MLPA are indicated

as red bars .

13

P40

(deletion of exons 31 to 47)

(deletion of exons 1 to 47)

P46 P53 P31 P49 P39 P34 P14 P20 P11 P09 P47

P32



27

4.1.1. Mutations causing premature stop codons

Frame-shift mutation - A novel mutation was identified in patient P53, c.86delC

(p.Pro29Hisfs*18), a single base deletion causing frame-shifting, and creating a stop

codon located 18 codons downstream (figure 4.3).

Figure 4.3: Sequencing electropherogram of a novel small

delet ion causing frame-shif t located in exon 3 of pat ient P53.

Nonsense mutations – Two nonsense mutations were also detected, c.1885C>T

(p.Arg629*) in exon 10 (patient P31), and c.3316C>T (p.Arg1106*) in exon 12 (patient

P20).

In patient P31 somatic mosaicism was suspected, since the c.1885C>T mutation is

underrepresented as compared with the wild type allele (figure 4.4A). Due to the lower

representation of the mutated allele, confirmation was made using a second

independent technique, namely high-resolution melting curve analysis (hrMCA) (Figure

4.4B). In addition, this change was not detected in 150 anonymized control samples.

c.86delC

Reference sequence

Foward sequence

Reverse sequence



28

4.1.2. In-frame deletion

Only one predictably in-frame deletion, c.6653_6655delATA, was found in P14.

This mutation comprises the deletion of the last two bases of an AAT codon

(asparagine), and the first base of the following ATT codon (isoleucine). This deletion

does not cause a shift in the reading frame, resulting in a deletion of a single aminoacid

(p.Asn2218del).

4.1.3. Missense mutations

Five mutations predicted to be missense were identified. Three were previously

reported (in P39, P34, and P09), namely: c.4422G>T (p.Arg1474Ser) in exon 21,

c.5471C>T (p.Ser1824Leu) in exon 29, and c.7168G>A (p.Ala2390Thr) in exon 42.

The other two mutations are novel, the mutation in patient P11, c.6983C>G

(p.Thr2328Arg) a transversion in exon 41 (Figure. 4.5A), and the mutation the

Figure 4.4: A) Suspected mosaicism for mutat ion c.1885C>T in exon 10, of NIPBL of

pat ient P31, detected by direct genomic sequencing. B) This mutat ion was conf irmed

through high resolut ion melt ing curve analysis. The normalized curve is on top right, and

dif ference graph on bottom right. Control samples are shown in blue, and pat ient ’s results

are in red.

c.1885C>T

Reference sequence

Foward sequence

Reverse sequence

A

B



29

c.7307C>T (p.Ala2436Val), a transition in exon 43 detected in patient P47 (Figure

4.5B).

Figure 4.5 : Sequencing electropherograms of two missense mutat ions; A) pat ient P11; B)

pat ient P47

In order to validate the pathogenicity of these variants a population screening was

performed by hrMCA. Both variants were not detected in at least 150 normal controls.

Further patogenicity assessment was carried out using different bioinformatic

algorithms (Table 4.2).

Both the p.Thr2328Arg and p.Ala2436Val mutations affect highly conserved

residues (Figure 4.6) located in the Armadillo-type fold protein domain.

In addition, the first mutation has a PolyPhen-2 score of 1.000 (probably

damaging mutation), and a SIFT score of 0.0 (deleterious), with predicted affected

protein function, and a moderate physicochemical difference between Threonine and

Arginine (Grantham distance of 71).

The p.Ala2436Val mutation has a PolyPhen-2 score of 0.977 (probably damaging

mutation), a SIFT score of 0.0 (deleterious), with a predicted affected protein function,

and a small physicochemical difference between Alanine and Valine (Grantham

distance of 64).

c.6983C>G

(p.Thr2328Arg)

Reference sequence

Foward sequence

Reverse sequence

A B

c.7307C>T

(p.Ala2436Val)



30

Table 4.2 : Evaluat ion of missense variant in terms of pathogenicity

a) Score ranging from 0 to 215, with higher value indicates larger difference; b) In the population screening, N/Nt -

indicates the number of control chromosomes with variant / total number of chromosomes tested.

Figure 4.6 : Protein sequence al ignments of delagin from several organisms (orthologs).

Affected residues, A) p.Thr2328Arg) and B) p.Ala2436Val are indicated by boxes.

Patient Mutation PolyPhen

v2 Score SIFT Score

Grantham

distance a)

Population

screening b)

P39 p.(Arg1474Ser) 0.534 0.0 (deleterious) 110 -

P34 p.(Ser1824Leu) 0.987 0.0 (deleterious) 145 -

P11 p.(Thr2328Arg) 0.998 0.0 (deleterious) 71 0/300

P09 p.(Ala2390Thr) 0.981 0.0 (deleterious) 58 -

P47 p.(Ala2436Val) 0.954 0.0 (deleterious) 64 0/300

A

B



31

The bioinformatic analysis results support the pathogenicity of the three previously

described mutations, and suggests that the two novel mutations are also pathogenic.

4.1.4. Splicing mutations

Two splicing mutations were identified: c.64+1G>A (P46), located in the canonical

sequence of the donor splice site of intron 2, and c.6763+5G>T (P20), a substitution on

intron 39 predicted to affect the donor splice site located upstream. Since no RNA

samples were available from patients, the impact of mutations at the expression level

was carried out by bioinformatic analysis using several algorithms that evaluate splicing

(Figures. 4.7 and 4.8).

Figure 4.7: Evaluat ion of P46 spl ic ing mutat ion c.64+1G>A using Alamut 3.0 ( Interact iv e Biosoftware, Rouen, France). On the top is displayed the reference sequence and on the bottom the mutated sequence. The column on the left shows f ive dist inct spl ice site predict ion methods (described in 3.9.2) and the grayed numbers to i ts r ight are the respect ive value ranges for each method score. Exons are shown in blue background. Score values are displayed as blue bars for 5 ’ donor sites, and green for 3’ acceptor si tes. Only scores that dif fer between the reference and mutated sequence are highl ighted in the f igure. The high bioinformatic predict ion scores for the 5’ donor site vanish completely from the mutated sequence, indicat ing a probable loss of the donor site with the mutat ion.

c.64+1G>A



32

Figure 4.8: Evaluat ion of pat ient P20 spl ic ing mutat ion, c.6763+5G>T using Alamut 3.0 (Interact ive Biosoftware, Rouen, France) . On the top is displayed the reference sequence and on the bottom the mutated sequence. The column on the left shows f ive dist inct spl ice site predict ion methods (described in 3.9.2) and the grayed numbers to i ts r ight are the respect ive value ranges for each method score. Exons are shown in blue boxes. Score values are displayed as blue bars for 5’ donor sites, and green for 3’ acceptor si tes. Only scores that dif fer between th e reference and mutated sequence are highl ighted in the f igure. Most of the high bioinformatic predict ion scores for the 5’ donor site disappear completely from the mutated sequence, indicat ing a probable loss of the donor site with the mutat ion .

4.1.5. Variant with unknown significance

A novel synonymous substitution found in exon 10, c.2727T>C (p.Gly909Gly)

(Figure 4.9A) was identified in patient P44. Population screening was carried out by

hrMCA, in 150 anonymized control samples in order to determine its frequency within

our population, but no mach was found (Figure 4.9B). Patient RNA was extracted and

converted to cDNA (3.2). The cDNA was amplified by PCR and sequenced (Figure

4.10). No alterations were found that suggested a splicing defect, moreover

sequencing results show a 1:1 ratio between normal and mutated allele compatible

with biallelic expression.

c.6763+5G>T



33

.

4.2. Large mutations detected in NIPBL gene

This screening was performed in all patients that were negative in the NIPBL gene

screening for point mutations and small deletions/insertions as previously (see 4.1)

Figure 4.11: Screening of NIPBL mutat ions by MLPA.

c.2727T>C

Figure 4.9: Variat ion detected in pat ient P44. A) Electropherogram of synonymous subst i tut ion

c.2727T>C on exon 10 B) Populat ion screening for c.2727T>C variant, through High Resolut ion

Melt ing, results are displayed on normalized curve (B) and dif ference graph (C). Control samples

are in blue, and the pat ient samples in red.

Figure 4.10 : cDNA electropherogram of P44 synonymous subst i tut ion on exon 10. Forward

sequence on the left , and reverse sequence on the right.

Reference sequence

Foward sequence

Reverse sequence

A

B

C



34

Results were analysed on GeneMarker® software using a statistical linear

regression model. Two novel gross rearrangements were found,

c.(5710-?)_(*927_?)del, a deletion spanning exons 31 to 47 detected in patient P32

(Figure 4.12 A and B), and c.(?_-481)_(*927_?)del, a deletion covering the entire

NIPBL gene (Figure 4.12 C and D) detected in patient P40.

Figure 4.12 : MLPA results of pat ient P32 (A and B, delet ion of exons 31 to 47) and pat ient P40

(C and D, delet ion of al l NIPBL exons). The 99% confidence level of the regression is shown within

the green l ines. Probes with lower rat ios (deleted) are shown as red squares outside the green

l ines. The green squares represent the normal DNA copy numbers and the blue squares the cont rol

probes. Standard deviat ion is shown in top left corner of each plot. A and C) MLPA probe set P141

(odd number exons). B) and D) MLPA probe set P142 (even number exons).

4.3. Mutation screening for SMC1A gene

This screening was made in patients previously tested with negative results for

mutations in the NIPBL gene. Since only a low percentage CdLS patients have been

reported with mutations in this gene (~5%), and because this gene has a limited

number of polymorphism and small exons (most exons are between 100 and 250 bp,

which allowed the design of small amplicons feasible for hrMCA technique), a different

approach, faster and cost-effective was used to screen SMC1A for mutations.

A B

C D



35

Figure 4.13: Strategy for SMC1A mutation screening by hrMCA following confirmation by sequencing.

An initial screening by high resolution melting curve analysis was done. In each run

three replicates for each sample and at least four negative controls were, in order to

minimize the possibility of inconsistent results which could result in false

positive/negative. After this screening, any sample that presented a shift/change on the

melting profile was sequenced to test for variants.

Three cases (not included in this study) presented differences from the normal

melting curve in several amplicons, but after being sequenced presented no variance in

the DNA sequence. After close analysis, it was detected that their melting curve

divergence was similar to samples extracted from blood with Lithium Heparin

anticoagulant addictive. Since samples in extracted with this addictive could not be

compared in with the control samples extracted from blood with K3-EDTA

anticoagulant addictive, they were removed from this study.

Figure 4.14 : Normalized High Resolut ion Melt ing Curves, for amplicons of exons 1 to 6a (exon

6 had to be ampli f ied in two amplicons due to i ts larger size). In Y axis is represented the relat ive

f luorescence of the samples (from 0 to 100), and on X axis temperature .

exon 1

exon 4

exon 3

exon 5 exon 6a

exon 2



36

Figure 4.15 : Normalized hrMCA prof i les for amplicons of exons 6 to 25. In Y axis is represented

the relat ive f luorescence of the samples (from 0 to 100), and on X axis the sample temperature.

exon 9 exon 11

exon 12

exon 22 exon 23

exon 24 exon 25

exon 6 (b) exon 8 exon 7

exon 13 exon 14

exon 15 exon 16 exon 17

exon 18 exon 19 exon 20

exon 21

exon 10



37

The melting curves obtained for each amplicon are shown below (figure 4.14 and

4.15). To validate this technique for diagnosis, samples that turned out negative for

variations in each of the amplicons were previously sequenced, to discard false

negatives. This sequencing is still ongoing, but at least five different samples for each

amplicon were sequenced, and no false negative result has appeared so far.

Using hrMCA as a screening method one variation was found in exon 9 of SMC1A

of patient P17 (Figures 4.15 and 4.16). Upon DNA sequencing this shift on the melting

profile was confirmed as being a true sequence variant: c.1487G>A, a previously

reported missense mutation (p.Arg496His), first reported by Deardorff et al., (2007)

(Figure 4.16).

Figure 4.16: SMC1A missense mutat ion, detected through High

Resolut ion Melt ing (A), (normalized curve on top, and dif ference graph at

bottom). Control samples are in blue, and the pat ient samples in red. (B)

Sequencing electropherogram of pat ient sample conf irming the presence

of a sequence variant (mutat ion c.1487G>A).

c.1487G>A

B

A

Reference sequence

Foward sequence

Reverse sequence



38

5. Discussion

5.1. Mutations identified in NIPBL

As shown in the results section this study includes the molecular characterization

of thirteen patients with NIPBL mutations. Eleven cases had sufficient clinical data

allowing their clinical classification: 3 (27%) were severe, 3 (27%) moderate and 5

(46%) have a milder CdLS phenotype. Considering the entire Portuguese CdLS cohort

(additional thirteen CdLS cases reported by our group in 2010) only five patients in a

total of 26 were classified as severe. This numbers contrast with the overall

percentages reported in the literature, where 40% of the CdLS cases were classified as

severe (Oliveira et al., 2010). Perhaps, this may be partially explained by the high

coverage of routine ultrasound scans during pregnancy in Portugal. The use of this

non-invasive prenatal screening technique has been proven very successful to prevent

several developmental defects, including upper limb reductions commonly associated

with the severest forms of CdLS. It is conceivable that these cases are identified by

prenatal scans resulting in medical termination of pregnancies, which in turn

contributes to lower the prevalence of the severe form of the disease.

In this work three novel mutations predicted to have the most drastic effects on the

protein were identified in patients with a severe phenotype. In the first case a complete

deletion of NIPBL gene was identified in patient P40. The second patient with a severe

phenotype (P2) has a large deletion that encompasses a significant part of NIPBL

(from exons 31 to 47); if efficient translation occurs this may lead to an aberrant

polypeptide with loss of part of the armadillo domain. The third patient (P53) has a

small frame-shift deletion (c.86del) that prematurely truncates the protein, where the

predicted ORF reading frame is limited to 47 codons. In this subset of our CdLS cohort,

the nonsense mutation (p.Arg1106*) identified in patient P49 was the only apparent

exception to this genotype-phenotype correlation. This mutation should be further

analyzed at the mRNA level, to exclude possible effects on splicing, such as exon

skipping or activation of a cryptic splice site. Some of these splicing defects lead to an

in-frame deletion at the protein level thereby explaining the milder form of CdLS.



39

Figure 5.1 : Representat ion of the NIPBL gene with the distribut ion of al l known point and small del/ ins mutat ions throughout the gene. The

blue rectangles represent the exons, and the white rectangles represent the int rons, and the blue rectangles The i ntronic region is are not up to

scale, being much larger than represented especial ly intron one. The pale blue rectangles are the non-coding exonic areas of the gene. Numbers

under the exons indicate the odd exons. Each red bar marked above represents the mutat ion frequency along the gene.

5’ 3’



40

The eleven point mutations detected in the NIPBL seem to be spread along the

gene without an apparent mutational hotspot, although their distribution seems non-

random. The majority of the reported NIPBL mutations are located in regions of the

gene with a higher mutation frequency (Figure 5.1) (Oliveira et al., 2010). A probable

explanation to this is the presence of gene specific sequence context such as

hypermutable CpG dinucleotides that might be affected by methylation-mediated

deamination (Cooper and Youssoufian, 1988), regions of repetitive sequences (like

short tandem repeats) and/or homopolymeric nucleotide stretches that are susceptible

to replication errors. In fact, the mutation detected on patient P14

(c.6653_6655delATA) is one of the most frequent recurrent mutations of the NIPBL

gene, being also an example of a mutation occurring in a repetitive sequences stretch

(deletion of one of three ATA repeats). In addition, five single substitutions c.64+1G>A

(patient P46), c.1885C>T (patient P10), c.3316C>T (patient P49), c.5471C>T (patient

P34), c.7168G>A (patient P09) are located in CpG dinucleotides. Seven mutations

including five missense mutations, an in-frame deletion and a change affecting splicing

were found inside the armadillo domain of delangin. This protein domain occupies a

considerable part of the gene (exons 13 to 37) and it is located in a highly conserved

region (Jahnke et al., 2008).

5.1.1. Detection of large NIPBL deletions by MLPA

Two large deletions in the NIPBL gene were found in our patients:

c.(?_-481)_(*927_?)del and c.(5710-?)_(*927_?)del. Although, CdLS diagnosis has

been confirmed, further molecular characterization should be made to determine the

breakpoints of these mutations. In the case of the patient with a complete deletion of

NIPBL (P40), this delineation would be of clinical importance to confirm or exclude the

involvement of neighboring genes. However, the patient has decease and there are no

additional biological samples available for analysis. The detection of two NIPBL gross

deletions in 3.8% of our patients (2/53) is quite relevant as compare to other CdLS

genes detection rates (please see section 5.3). It shows the importance of using MLPA

technique after exclusion of point mutations by genomic sequencing in order to

increase the overall NIPBL mutation detection rate. These numbers are very similar to

those reported in a recent study, where the same strategy detected mutations in 3.5%

of patients (7/200) (Russo et al., 2012). Overall, these results confirm the



41

recommendation for the use of MLPA technique on NIPBL in a routine molecular

diagnostic setting.

5.1.2. Somatic mosaicism

As it was shown in the results section, the c.1885C>T mutation (p.Arg629*) in exon

10 of patient P31 was suspected of somatic mosaicism. Compared to the wild type (wt)

the mutated allele is under-represented, <25% of the fluorescence intensity of the wt

allele, estimated by comparing both signals in the sequence electropherogram. This

mutation if not carefully looked-up could have been missed during analysis using

default settings. Myake et al. (2005) reported a Japanese patient having this nonsense

mutation associated to a severe phenotype; while patient P31 was classified as

moderate. This discrepancy is also in favor of mosaicism, since patients with mosaic

mutations tend to express an attenuated phenotype than non-mosaic patients with the

same mutation (Kluwe et al., 2003). For further characterization of this case

quantitative methods such as real-time PCR should be applied to quantify the

percentage of cells with the mutation as compared to wt, ideally this quantification

should be perform on DNA obtained from different tissues (from distinct germ layers).

Mendelian disorders with a high frequency of de novo mutations (as is the case of

CdLS) have a relatively frequent occurrence of somatic mosaicism (Castronovo et al.,

2010). However very few of such cases have been reported in NIPBL so far

(http://www.lovd.nl/NIPBL, last accessed September 2012). It is possible that some

somatic mutations with low-level mutant alleles could escape detection by routine

sequencing methods. Somatic mosaicism in NIPBL gene could help to explain the

phenotypic variability of patients, and is another source of clinical and genetic

heterogeneity of this syndrome (Castronovo et al., 2010). These observations raise the

possibility that mutation rates of genes involved in CdLS are currently underestimated.

In addition, as demonstrated by the use of hrMCA technique it is important to combine

different techniques for diagnosis purposes. In molecularly unresolved cases next

generation sequencing (NGS) would be a powerful tool to detect pathogenic mosaic

mutations since this approach, specially using a high coverage strategy, is sensible

enough to detect low allele frequencies (Pagnamenta et al., 2012). Furthermore, Next

generation sequencing (NGS) allows the screening of several genes simultaneously

which is another advantage for its use in this field.

http://www.ncbi.nlm.nih.gov/pubmed?term=Pagnamenta%20AT%5BAuthor%5D&cauthor=true&cauthor_uid=22129557



42

5.1.3. Novel silent polymorphism

The apparently silent substitution c.2727T>C sequence variant in NIPBL detected

in patient P44 was further studied at the mRNA level. We have shown that this change,

although it was not detected in control samples, does not have an impact on splicing

(inclusion of exon 10). Very recently this variant was reported in the Single Nucleotide

Polymorphism database of NCBI (dbSNP, rs148394805), as a rare variant with an

allelic frequency of only 0.2%. Due to its very low frequency it was not detected during

our population screening.

5.2. Implementation of hrMCA for SMC1A

One of the goals of this work was to set up a molecular diagnostic for mutation

screening in SMC1A gene. The fast, efficiency and low-cost of the hrMCA technique,

combined with the gene characteristics (small easily amplifiable exons and limited

number of polymorphisms) made this technique an ideal choice for the mutation

screening of SMC1A. This is the first known study to apply hrMCA technique for

mutation screening on SMC1A gene. While there are no references in the literature for

the use of the technique in this gene, studies in other genes stated the robustness,

high-throughput, high sensitivity and low false positive ratio of this technique (Millat et

al., 2009; Santos et al., 2012). Because of its high sensitivity the melting curve profile

can be altered by several secondary factors like the presence of primer-dimers,

amplification of non specific products, or using DNA templates with different

concentrations or quality/purity. These might cause false positives results but it can be

easily avoided by a good PCR optimization. The use of at least three replicas for each

sample and the inclusion of previously genotyped controls may avoid problems with

individual samples. Amplicons that present a shifted melting curve have to be

sequenced. Although, this technique is extremely sensitive to detect a variation, it

cannot determine what the precise sequence change is unless positive controls are

added to the same run. DNA sequencing after hrMCA is also useful to exclude false

positives. hrMCA scanning accuracy depends on high quality PCR (Santos et al., 2012),

but as long as optimal PCR conditions can be assured, and if combined with genome



43

sequencing to validate results, this technique can provide an fast and effective way of

screening SMC1A gene for mutations.

The proof-of-principle of the successful deployment of this technique for mutation

scanning in SMC1A was the identification of a sequence variant (c.1487G>A) in patient

P17, initially detected by hrMCA and later confirmed by genomic sequencing. This

missense mutation (p.Arg496His) identified in exon 9 of SMC1A, was previously

reported in six other CdLS patients (Deardorff et al., 2007). The mutation occurs in the

SMC flexible hinge domain. A previous study reported that mutations in this domain

increase the affinity for DNA binding, comparing with the wild type proteins (Revenkova

et al., 2009).

5.3. Mutation detection rate

The Portuguese CdLS patient cohort studied by our group includes a total of fifty-

three individuals studied by different techniques applied to SMC1A and NIPBL genes.

Overall results have allowed the successful molecular characterization of twenty-seven

patients, the majority with mutations in NIPBL gene and only one case with a

pathogenic variant in SMC1A (Table 5.1).

The mutation detection rate for NIPBL gene is 49.1% (26/53) which is within the

range of values (40-60%) indicated in the literature for this locus (Russo et al., 2012).

Although some research groups report mutation detection rates for NIPBL gene up to

60%. This might be attributed to different inclusion or exclusion criteria for selection of

patients that could lead for the mildest phenotypes to be excluded from NIPBL gene

analysis.

Concerning SMC1A gene, the mutation detection rate was 1.9% below the 5%

detection rate reported in the literature by other groups (Russo et al., 2012). One

possible explanation may be attributed to the fact that SMC1A gene was not completed

sequenced in all NIPBL-negative cases. It is conceivable that some mutations may

have not been detected by hrMCA that was used as a prospective scanning method.

SMC1A sequencing analysis should be performed in all molecularly unresolved cases

to exclude the involvement of this locus.



44

Table 5.1 : Overal l results of the genetic analysis of NIPBL and SMC1A genes in the Portuguese

CdLS pat ient cohort.

SEQ- genomic sequencing; MLPA- multiplex ligation probe amplification; HRM- high

resolution melting; P- patients; * patients described in a previous publication (Oliveira et al.,

2010).

5.4. Bioinformatic tools and databases

Nowadays, bioinformatics tools are essential resources in a clinical molecular

genetics laboratory. From primer design, to sequence analysis and mutation

interpretation these applications speed up work and provide useful additional data for

mutation validation. During this work several in-silico tools were used to validate new

missense changes and mutations that affected splicing. Unfortunately, in some of the

patients it was not possible to obtain new biological samples essential to perform

additional studies, so the use of bioinformatic prediction tools can help to ameliorate

these limitations.

The continuous build up of new sequence variant led the international scientific

community to create breakthrough initiatives such as the Human Variome Project

(HVP) (Kohonen-Corish et al., 2010). One of the main objectives of HVP is the creation

of LSDBs for all Human genes implicated in mendelian diseases. The development and

update of such databases are greatly dependent on the activities of several research

groups. The use of mutation databases like the LSDBs for NIPBL and SMC1A, are an

efficient way of sharing information regarding variants and phenotypes. Since obtaining

evidence of pathogenicity of variants is costly and time-consuming these databases are

Patients

Mutation positive CdLS cases

Negative cases NIPBL

(SEQ) NIPBL (MLPA)

SMC1A (HRM)

Total

Phase 1 (P1-39)

39 20

1 1 22 17 (7+13*)

Phase 2 (P40-53)

14 4 1 0 5 9

Total 53 24

(11+13*) 2 1 27 26



45

also important resources for diagnostic laboratories, which can rapidly cross-check if a

sequence changes in one of the CdLS genes have been already reported.

5.5. Procedure for the molecular diagnostic of CdLS patients

Due to the genetic complexity of CdLS which concerns different mutation types and

multiple genes, a procedure flowchart for the genetic molecular diagnostic of CdLS is

proposed (Figure 5.2). The sequential order of the different techniques was established

considering their sensitivity to optimize time and economical resources. The first gene

that should be considered for analysis is NIPBL since the majority of mutations

described up to now are related to this locus. Direct genomic sequence should be the

first technique to be applied, considering that the most frequent variants are point

mutations or small deletions/insertions. Although it was not included in the flowchart, it

is of importance to emphasize that approximately 80% of NIPBL mutations described to

date are located on 21 exons (out of a total of 47) (Oliveira et al., 2010). Therefore,

NIPBL sequencing step may be sub-divided in two steps, starting by exons with higher

mutation frequency and afterwards sequencing the remaining 26 exons of NIPBL with

less mutations reported. Theoretically, this partial NIPBL gene sequencing would allow

detection of mutations in about one third of the CdLS patients. After NIPBL genomic

sequencing, cases still unresolved should be submitted to MLPA technique to detected

large deletions or duplication involving one or more exons of NIPBL. As discussed

above, this technique has a greater mutation detection rate than sequencing SMC1A

itself. Results have shown that hrMCA enables a fast and cost-effective scanning of

SMC1A. However, hrMCA technique must always be validated by genomic sequencing

to exclude false positives and polymorphisms.

As mentioned in the Introduction, besides NIPBL and SMC1A other genes, to be

precise SMC3, RAD21, and HDAC8, have been associated with milder CdLS (or

Cornelia-like) phenotypes in a limited number of cases (Revenkova et al., 2009;

Deardorff et al., 2012a). In spite of the low mutation detection rate, these genes should

be considered for genetic analysis, especially if there is a specific clinical request

based upon characteristic phenotypic traits or suggestive patterns of inheritance (such

as X-linked).



46

Figure 5.2: Flowchart for a proposed procedure for the genetic molecular diagnost ic of CdLS



47

5.6. Future perspectives

All the available data within public domain suggests that at least 30% of CdLS

patients are still molecularly unsolved. This prompts to new lines of research dedicated

to the identification of new genetic causes for CdLS. The obvious candidates can be

found among the genes involved with the cohesion complex and their regulators. A

new technological paradigm is currently changing the face of Human Genetics, next

generation sequencing (NGS), which consist in extremely high-throughput genetic

analyzers. NGS technology can be used for CdLS research by at least two different

approaches. The first is based on the development of custom genomic libraries for

targeted resequencing where a number of candidate genes (such as those involved in

the cohesin complex) are selected and analyzed simultaneously in several patients. An

alternative strategy is in well selected CdLS patients perform a whole exome

sequencing, meaning that all the coding regions of all known Human genes are

sequenced to identify variants responsible for CdLS.



48

6. Bibliography

Adzhubei IA, Schmidt S, Peshkin L, Ramensky VE, Gerasimova A, Bork P,

Kondrashov AS, Sunyaev SR. (2010). A method and server for predicting damaging

missense mutations. Nature Methods. 7(4):248-249.

Bhuiyan ZA, Stewart H, Redeker EJ, Mannens MM, Hennekam RC. (2007). Large

genomic rearrangements in NIPBL are infrequent in Cornelia de Lange syndrome.

European Journal of Human Genetics. 15(4):505-508.

Borck G, Redon R, Sanlaville D, Rio M, Prieur M, Lyonnet S, Vekemans M, Carter

NP, Munnich A, Colleaux L, Cormier-Daire V. (2004). NIPBL mutations and genetic

heterogeneity in Cornelia de Lange syndrome. Journal of Medical Genetics.

41(12):e128.

Borck G, Zarhrate M, Bonnefont JP, Munnich A, Cormier-Daire V, Colleaux L.

(2007). Incidence and clinical features of X-linked Cornelia de Lange syndrome due to

SMC1L1 mutations. Human Mutation. 28(2):205-206.

Braunholz D, Hullings M, Gil-Rodríguez MC, Fincher CT, Mallozzi MB, Loy E,

Albrecht M, Kaur M, Limon J, Rampuria A, Clark D, Kline A, Dalski A, Eckhold J,

Tzschach A, Hennekam R, Gillessen-Kaesbach G, Wierzba J, Krantz ID, Deardorff MA,

Kaiser FJ. (2012). Isolated NIBPL missense mutations that cause Cornelia de Lange

syndrome alter MAU2 interaction. European Journal of Human Genetics. 20(3):271-

276.

Buggy JJ, Sideris ML, Mak P, Lorimer DD, McIntosh B, Clark JM. (2000). Cloning

and characterization of a novel human histone deacetylase, HDAC8. Biochemical

Journal. 350 Pt 1:199-205.

Burpo FJ. (2001). A critical review of pcr primer design algorithms and cross-

hybridization case study. Biochemistry 2001:218-233

Carramolino L, Lee BC, Zaballos A, Peled A, Barthelemy I, Shav-Tal Y, Prieto I,

Carmi P, Gothelf Y, González de Buitrago G, Aracil M, Márquez G, Barbero JL, Zipori

D. (1997). SA-1, a nuclear protein encoded by one member of a novel gene family:

molecular cloning and detection in hemopoietic organs. Gene. 195(2):151-159.



49

Castronovo P, Delahaye-Duriez A, Gervasini C, Azzollini J, Minier F, Russo S,

Masciadri M, Selicorni A, Verloes A, Larizza L. (2010). Somatic mosaicism in Cornelia

de Lange syndrome: a further contributor to the wide clinical expressivity? Clinical

Genetics. 78(6):560-564.

de Lange, C. (1933). Sur un type nouveau de dégénération (Typus

Amstelodamensis). Médecine des Enfants. XXXVI-12:714-719.

Deardorff MA, Kaur M, Yaeger D, Rampuria A, Korolev S, Pie J, Gil-Rodríguez C,

Arnedo M, Loeys B, Kline AD, Wilson M, Lillquist K, Siu V, Ramos FJ, Musio A,

Jackson LS, Dorsett D, Krantz ID. (2007). Mutations in cohesin complex members

SMC3 and SMC1A cause a mild variant of cornelia de Lange syndrome with

predominant mental retardation. American Journal of Human Genetics. 80(3):485-494.

Deardorff MA, Clark DM, Krantz ID. (2011). Cornelia de Lange Syndrome. Gene

Review. University of Washington, Seattle. Last accessed online on 27 September

2012 on: http://www.ncbi.nlm.nih.gov/books/NBK1104/.

Deardorff MA, Bando M, Nakato R, Watrin E, Itoh T, Minamino M, Saitoh K,

Komata M, Katou Y, Clark D, Cole KE, De Baere E, Decroos C, Di Donato N, Ernst S,

Francey LJ, Gyftodimou Y, Hirashima K, Hullings M, Ishikawa Y, Jaulin C, Kaur M,

Kiyono T, Lombardi PM, Magnaghi-Jaulin L, Mortier GR, Nozaki N, Petersen MB,

Seimiya H, Siu VM, Suzuki Y, Takagaki K, Wilde JJ, Willems PJ, Prigent C, Gillessen-

Kaesbach G, Christianson DW, Kaiser FJ, Jackson LG, Hirota T, Krantz ID, Shirahige

K. (2012a). HDAC8 mutations in Cornelia de Lange syndrome affect the cohesin

acetylation cycle. Nature. 489(7415):313-317.

Deardorff MA, Wilde JJ, Albrecht M, Dickinson E, Tennstedt S, Braunholz D,

Mönnich M, Yan Y, Xu W, Gil-Rodríguez MC, Clark D, Hakonarson H, Halbach S,

Michelis LD, Rampuria A, Rossier E, Spranger S, Van Maldergem L, Lynch SA,

Gillessen-Kaesbach G, Lüdecke HJ, Ramsay RG, McKay MJ, Krantz ID, Xu H,

Horsfield JA, Kaiser FJ. (2012b). RAD21 mutations cause a human cohesinopathy.

American Journal of Medical Genetics. 90(6):1014-1027.

Desmet FO, Hamroun D, Lalande M, Collod-Beroud G, Claustres M, Beroud C.

(2009) Human Splicing Finder: an online bioinformatics tool to predict splicing signals.

Nucleic Acid Research, 37(9):e67.

http://www.ncbi.nlm.nih.gov/pubmed?term=Clark%20DM%5BAuthor%5D&cauthor=true&cauthor_uid=20301283

http://www.ncbi.nlm.nih.gov/pubmed?term=Krantz%20ID%5BAuthor%5D&cauthor=true&cauthor_uid=20301283

http://www.ncbi.nlm.nih.gov/pubmed?term=Cornelia%20de%20Lange%20Syndrome%3A%20Clinical%20Review%2C%20Diagnostic%20and%20Scoring%20Systems%2C%20and%20Anticipatory%20Guidance

http://www.ncbi.nlm.nih.gov/pubmed?term=Cornelia%20de%20Lange%20Syndrome%3A%20Clinical%20Review%2C%20Diagnostic%20and%20Scoring%20Systems%2C%20and%20Anticipatory%20Guidance



50

Dias C., Costa E., Oliveira J., Silva J., Martins M., Fortuna A. M., Sousa A. B.,

Basto J. P., Soares-Silva I., Kay T., Santos H., Medeira A., Cordeiro I., Lourenço T.,

Nunes L., Santos R., Reis-Lima M. (2008). Clinical characterization and NIPBL

mutation analysis of 42 Portuguese patients with Cornelia de Lange Syndrome.

Barcelona, May 2008. European Society of Human Genetics. 16-2:97

Dorsett D. (2007). Cohesin, gene expression and development: lessons from

Drosophila. Chromosome Research. 17(2):185-200.

Dorsett D, Krantz ID. (2009). On the molecular etiology of Cornelia de Lange

syndrome. Annals of the New York Academy of Sciences. 1151:22-37.

Fay A, Misulovin Z, Li J, Schaaf CA, Gause M, Gilmour DS, Dorsett D. (2011).

Cohesin selectively binds and regulates genes with paused RNA polymerase. Current

Biology. 11;21(19):1624-1634.

Galehdari H, Monajemzadeh R, Nazem H, Mohamadian G, Pedram M. (2011).

dentification of a novel de novo mutation in the NIPBL gene in an Iranian patient with

Cornelia de Lange syndrome: A case report. Journal of Medical Case Reports. 5:242.

Geck P, Maffini MV, Szelei J, Sonnenschein C, Soto AM. (2000). Androgen-

induced proliferative quiescence in prostate cancer cells: the role of AS3 as its

mediator. Proceedings of the National Academy of Sciences USA. 97(18):10185-10190.

Glasel, JA. (1995). Validity of Nucleic Acid Purities Monitored by A260/A280

Absorbance Ratios. Biotechniques 18-1:62-63.

Gordillo M, Vega H, Trainer AH, Hou F, Sakai N, Luque R, Kayserili H, Basaran S,

Skovby F, Hennekam RC, Uzielli ML, Schnur RE, Manouvrier S, Chang S, Blair E,

Hurst JA, Forzano F, Meins M, Simola KO, Raas-Rothschild A, Schultz RA, McDaniel

LD, Ozono K, Inui K, Zou H, Jabs EW. (2008). The molecular mechanism underlying

Roberts syndrome involves loss of ESCO2 acetyltransferase activity. Human Molecular

Genetics. 17(14):2172-2180.

Hoppman-Chaney N, Jang JS, Jen J, Babovic-Vuksanovic D, Hodge JC., (2011).

In-frame multi-exon deletion of SMC1A in a severely affected female with Cornelia de

Lange Syndrome. American Journal of Medical Genetics. Part A 158A:193-198.



51

Horsfield JA, Print CG, Mönnich M. (2012). Diverse developmental disorders from

the one ring: distinct molecular pathways underlie the cohesinopathies. Frontiers in

Genetics. 3:171-186.

Hou F, Zou H. (2005). Two human orthologues of Eco1/Ctf7 acetyltransferases are

both required for proper sister-chromatid cohesion. Molecular Biology of the Cell.

16(8):3908-3918.

Jahnke P, Xu W, Wülling M, Albrecht M, Gabriel H, Gillessen-Kaesbach G, Kaiser

FJ. (2008). The Cohesin loading factor NIPBL recruits histone deacetylases to mediate

local chromatin modifications. Nucleic Acids Research. 36(20):6450-6458.

Johns DA, Bhonsale DL, Shivashanker VY, Johns M. (2012). Aesthetic and

functional management of a patient with Cornelia de Lange syndrome. Contemporary

Clinical Dentistry. 3(Suppl 1):S86-S91.

Kalendar R, Lee D, Schulman AH (2011). Java web tools for PCR, in silico PCR,

and oligonucleotide assembly and analysis. Genomics, 98(2): 137-144.

Kline AD, Krantz ID, Sommer A, Kliewer M, Jackson LG, FitzPatrick DR, Levin AV,

Selicorni A. (2007). Cornelia de Lange syndrome: clinical review, diagnostic and

scoring systems, and anticipatory guidance. American Journal of medical genetics

143A(12):1287-1296.

Kluwe L, Mautner V, Heinrich B, Dezube R, Jacoby LB, Friedrich RE, MacCollin M.

(2003). Molecular study of frequency of mosaicism in neurofibromatosis 2 patients with

bilateral vestibular schwannomas. Journal of Medical Genetics. 40(2):109-114.

Kohonen-Corish MR, Al-Aama JY, Auerbach AD, Axton M, Barash CI, Bernstein I,

Béroud C, Burn J, Cunningham F, Cutting GR, den Dunnen JT, Greenblatt MS, Kaput J,

Katz M, Lindblom A, Macrae F, Maglott D, Möslein G, Povey S, Ramesar R, Richards

S, Seminara D, Sobrido MJ, Tavtigian S, Taylor G, Vihinen M, Winship I, Cotton RG;

Human Variome Project Meeting. (2010). How to catch all those mutations--the report

of the third Human Variome Project Meeting, UNESCO Paris, May 2010. Human

Mutation 31(12):1374-1381.

Krantz ID, McCallum J, DeScipio C, Kaur M, Gillis LA, Yaeger D, Jukofsky L,

Wasserman N, Bottani A, Morris CA, Nowaczyk MJ, Toriello H, Bamshad MJ, Carey JC,

Rappaport E, Kawauchi S, Lander AD, Calof AL, Li HH, Devoto M, Jackson LG. (2004).



52

Cornelia de Lange syndrome is caused by mutations in NIPBL, the human homolog of

Drosophila melanogaster Nipped-B. Nature Genetics. 36(6):631-635.

Kueng S, Hegemann B, Peters BH, Lipp JJ, Schleiffer A, Mechtler K, Peters JM.

(2006). Wapl controls the dynamic association of cohesin with chromatin. Cell.

127(5):955-967

Kumar D, Blank CE, Griffiths BL. (1985). Cornelia de Lange syndrome in several

members of the same family. Journal of Medical Genetics, 22, 296-300.

Kumar D, Sakabe I, Patel S, Zhang Y, Ahmad I, Gehan EA, Whiteside TL, Kasid U.

(2004). SCC-112, a novel cell cycle-regulated molecule, exhibits reduced expression in

human renal carcinomas. Gene. 328:187-196.

Lalatta F, Russo S, Gentilin B, Spaccini L, Boschetto C, Cavalleri F, Masciadri M,

Gervasini C, Bentivegna A, Castronovo P, Larizza L. (2007). Prenatal/neonatal

pathology in two cases of Cornelia de Lange syndrome harboring novel mutations of

NIPBL. Genetics in Medicine. 9(3):188-194.

Liu J, Krantz ID. (2009). Cornelia de Lange syndrome, cohesin, and beyond.

Clinical Genetics. 76(4):303–314.

Maniatis T, Fritsch EF, and Sambrook J. (1982) Molecular Cloning A Laboratory

Manual, Cold Spring Harbor Laboratory, Cold Springs Harbor, NY.

Mannini L, Liu J, Krantz ID, Musio A. (2010). Spectrum and consequences of

SMC1A mutations: the unexpected involvement of a core component of cohesin in

human disease. Human Mutation. 31(1):5-10.

Millat G, Chanavat V, Julia S, Crehalet H, Bouvagnet P, Rousson R. (2009).

Validation of high-resolution DNA melting analysis for mutation scanning of the LMNA

gene. Clinical Biochemistry. 42(9):892-898

Miller SA, Dykes DD, and Polesky HF. (1998). A simple salting out procedure for

extracting DNA from human nucleated cells. Nucleic Acids Research. 16(3): 1215.

Musio A, Selicorni A, Focarelli ML, Gervasini C, Milani D, Russo S, Vezzoni P,

Larizza L. (2006). X-linked Cornelia de Lange syndrome owing to SMC1L1 mutations.

Nature Genetics. 8(5):528-530.



53

Miyake N, Visser R, Kinoshita A, Yoshiura K, Niikawa N, Kondoh T, Matsumoto N,

Harada N, Okamoto N, Sonoda T, Naritomi K, Kaname T, Chinen Y, Tonoki H,

Kurosawa K. (2005). Four novel NIPBL mutations in Japanese patients with Cornelia

de Lange syndrome. American Journal of Medical Genetics. 135(1):103-105.

Nishiyama T, Ladurner R, Schmitz J, Kreidl E, Schleiffer A, Bhaskara V, Bando M,

Shirahige K, Hyman AA, Mechtler K, Peters JM. (2010). Sororin mediates sister

chromatid cohesion by antagonizing Wapl. Cell. 143(5):737-749.

Oliveira J, Dias C, Redeker E, Costa E, Silva J, Reis Lima M, den Dunnen JT,

Santos R. (2010). Development of NIPBL locus-specific database using LOVD: from

novel mutations to further genotype-phenotype correlations in Cornelia de Lange

Syndrome. Human Mutation. 31(11):1216-1222.

Oostra RJ, Baljet B, Hennekam RC. Brachmann-de Lange syndrome "avant la

lettre". (1994). American Journal of Medical Genetics. 1994 Sep 1;52(3):267-268.

Pagnamenta AT, Lise S, Harrison V, Stewart H, Jayawant S, Quaghebeur G, Deng

AT, Murphy VE, Sadighi Akha E, Rimmer A, Mathieson I, Knight SJ, Kini U, Taylor JC,

Keays DA. (2012). Exome sequencing can detect pathogenic mosaic mutations

present at low allele frequencies. Journal of Human Genetics. 57(1):70-72

Parisi S, McKay MJ, Molnar M, Thompson MA, van der Spek PJ, van Drunen-

Schoenmaker E, Kanaar R, Lehmann E, Hoeijmakers JH, Kohli J. (1999). Molecular

and Cellular Biology. 19(5):3515-3528.

Pauline C. and Henikoff S, (2003) SIFT: predicting amino acid changes that affect

protein function. Nucleic Acid research. 31(13): 3812–3814.

Pezzi N, Prieto I, Kremer L, Pérez Jurado LA, Valero C, Del Mazo J, Martínez-A C,

Barbero JL. (2000). STAG3, a novel gene encoding a protein involved in meiotic

chromosome pairing and location of STAG3-related genes flanking the Williams-

Beuren syndrome deletion. Journal of the Federation of American Society for

Experimental Biology. 14(3):581-592.

Pertea M, Lin X, Salzberg SL. GeneSplicer: a new computational method for splice

site prediction. Nucleic Acid Research. 29(5):1185-1190.

http://www.ncbi.nlm.nih.gov/pubmed?term=Baljet%20B%5BAuthor%5D&cauthor=true&cauthor_uid=7810556

http://www.ncbi.nlm.nih.gov/pubmed?term=Hennekam%20RC%5BAuthor%5D&cauthor=true&cauthor_uid=7810556

http://www.ncbi.nlm.nih.gov/pubmed/7810556



54

Reese MG, Eeckman FH, Kulp D, Haussler D. (1997) Improved splice site

detection in Genie. 4(3):311-323.

Rankin S, Ayad NG, Kirschner MW. (2005). Sororin, a substrate of the anaphase-

promoting complex, is required for sister chromatid cohesion in vertebrates. Molecular

Cell. 18(2):185-200.

Revenkova E, Eijpe M, Heyting C, Gross B, Jessberger R. (2001). Novel meiosis-

specific isoform of mammalian SMC1. Molecular Cell Biology. 21(20):6984-6998.

Revenkova E, Eijpe M, Heyting C, Hodges CA, Hunt PA, Liebe B, Scherthan H,

Jessberger R. (2004). Cohesin SMC1 beta is required for meiotic chromosome

dynamics, sister chromatid cohesion and DNA recombination. Nature Cell Biology.

6(6):555-562.

Revenkova E, Focarelli ML, Susani L, Paulis M, Bassi MT, Mannini L, Frattini A,

Delia D, Krantz I, Vezzoni P, Jessberger R, Musio A. (2009). Cornelia de Lange

syndrome mutations in SMC1A or SMC3 affect binding to DNA. Human Molecular

Genetics. 18(3):418-427.

Russo S, Masciadri M, Gervasini C, Azzollini J, Cereda A, Zampino G, Haas O,

Scarano G, Di Rocco M, Finelli P, Tenconi R, Selicorni A, Larizza L. (2012). Intragenic

and large NIPBL rearrangements revealed by MLPA in Cornelia de Lange patients.

European Journal of Human Genetics. 20(7):734-741.

Sadano H, Sugimoto H, Sakai F, Nomura N, Osumi T. (2000) NXP-1, a human

protein related to Rad21/Scc1/Mcd1, is a component of the nuclear matrix. Biochemical

and Biophysical Research Communications. 267(1):418-422.

Santos S, Marques V, Pires M, Silveira L, Oliveira H, Lança V, Brito D, Madeira H,

Esteves JF, Freitas A, Carreira IM, Gaspar IM, Monteiro C, Fernandes AR. (2012) High

resolution melting: improvements in the genetic diagnosis of hypertrophic

cardiomyopathy in a Portuguese cohort. BMC Medical Genetics. 13:17.

Selicorni A, Russo S, Gervasini C, Castronovo P, Milani D, Cavalleri F, Bentivegna

A, Masciadri M, Domi A, Divizia MT, Sforzini C, Tarantino E, Memo L, Scarano G,

Larizza L. (2008). Clinical score of 62 Italian patients with Cornelia de Lange syndrome

and correlations with the presence and type of NIPBL mutation. Clinical Genetics.

72(2):98-108.



55

Sumara I, Vorlaufer E, Gieffers C, Peters BH, Peters JM. (2000). Characterization

of vertebrate cohesin complexes and their regulation in prophase. 151(4):749-762.

Tonkin ET, Wang TJ, Lisgo S, Bamshad MJ, Strachan T. (2004). NIPBL, encoding

a homolog of fungal Scc2-type sister chromatid cohesion proteins and fly Nipped-B, is

mutated in Cornelia de Lange syndrome, Nature Genetics. 36(6):636-641.

Vega H, Waisfisz Q, Gordillo M, Sakai N, Yanagihara I, Yamada M, van Gosliga D,

Kayserili H, Xu C, Ozono K, Jabs EW, Inui K, Joenje H. (2005). Roberts syndrome is

caused by mutations in ESCO2, a human homolog of yeast ECO1 that is essential for

the establishment of sister chromatid cohesion. 37(5):468-470.

Van den Wyngaert I, de Vries W, Kremer A, Neefs J, Verhasselt P, Luyten WH,

Kass SU. (2000). Cloning and characterization of human histone deacetylase 8. FEBS

Letters. 478(1-2):77-83.

Watrin E, Peters JM. (2006). Cohesin and DNA damage repair. Experimental Cell

Research. 312(14):2687-2693.

Watrin E, Schleiffer A, Tanaka K, Eisenhaber F, Nasmyth K, Peters JM. (2006).

Human Scc4 is required for cohesin binding to chromatin, sister-chromatid cohesion,

and mitotic progression. Current Biology. 16(9):863-874.

Yan J, Saifi GM, Wierzba TH, Withers M, Bien-Willner GA, Limon J, Stankiewicz P,

Lupski JR, Wierzba J. (2006). Mutational and genotype-phenotype correlation analyses

in 28 Polish patients with Cornelia de Lange syndrome. American Journal of Human

Genetics. 140(14):1531-1541.

Ye J, Coulouris G, Zaretskaya I, Cutcutache I, Rozen S, and Madden TL (2012).

Primer-BLAST: A tool to design target-specific primers for polymerase chain reaction.

Bioinformatics. 13:134.

Yeo G, Burge CB. (2004). Maximum entropy modeling of short sequence motifs

with applications to RNA splicing signals. Journal of Computational Biology. 11(2-

3):377-94.

Zhang B, Chang J, Fu M, Huang J, Kashyap R, Salavaggione E, Jain S, Kulkarni S,

Deardorff MA, Uzielli ML, Dorsett D, Beebe DC, Jay PY, Heuckeroth RO, Krantz I,



56

Milbrandt J. (2009). Dosage effects of cohesin regulatory factor PDS5 on mammalian

development: implications for cohesinopathies. PLoS One. 4(5):e5232.



I

Annex I Clinical Features used for Kline et al., 2007 scoring system

a) facial features – including synophrys with arched fine eyebrows, long

eyelashes, short nose, anteverted nares, long prominent philtrum, thin lips,

high palate

b) delayed growth – small stature, microcephaly

c) developmental delay – mental retardation, learning disabilities, speech and

language deficits)

d) behavioral issues – like attention deficit disorder, obsessive-compulsive

disorder characteristics, anxiety, aggression and self-injury, and autistic-like

features

e) musculoskeletal anomalies – extremities with reduction defects, arm and hand

anomalies, and chest, back, and hip complications

f) neurosensory and skin problems – including ophthalmic, auditory and central

nervous systems, hirsutism, cutis marmorata

g) other major systemic involvement – ex.: gastroesophageal reflux disease,

genitourinary anomalies, congenital heart disease)



II

Annex II CdLS genetic study request form



III

Annex III

Table of primers used in this study

a. NIPBL (Accession Number NM_133433.3)

a.1 M13-Tailed custom primers

Primer Sequence (5’ -> 3’)

Ta

annealing temperature

g.NIPBL-2-F g.NIPBL-2-F

TGTAAAACGACGGCCAGTAGGTTGAACAAACCAAAGCAGTAAC CAGGAAACAGCTATGACCAAGGGCAGTTTCAGTTGCTATTTC

58

g.NIPBL-3-F g.NIPBL-3-R

TGTAAAACGACGGCCAGTCCCAAAATACAGATAAGCACTAAAGAGAC CAGGAAACAGCTATGACCTTCTGAAATAAAACCAGGAATACGG

58


TGTAAAACGACGGCCAGTAAATTCATATTGTTGGCCATACC CAGGAAACAGCTATGACCCAAGAATCTTATCAGTTTAGTTTCCTTC

58

g.NIPBL-5-F g.NIPBL-5-F

TGTAAAACGACGGCCAGTACAGTTTTACAGCGTCTATATTTTGCTC CAGGAAACAGCTATGACCAATCCACCATTACCTAAATGTATGAACC

58


TGTAAAACGACGGCCAGTACGGAGCATGGCTTTAAGTACC CAGGAAACAGCTATGACCCTGAATCCAAAAAGCCAAAATG

58


TGTAAAACGACGGCCAGTTCTGGCAGAGTGTTAAGATTTGATG CAGGAAACAGCTATGACCCCTGCTCTAGACTAAGCTCTGTATGTG

58


TGTAAAACGACGGCCAGTTGCCATGCTTAGGCTGTTG CAGGAAACAGCTATGACCAAATCCCATGATCCAGACCTG

58

g.NIPBL-9-1-F g.NIPBL-9-1-R

TGTAAAACGACGGCCAGTAGCTAAACAACTGTGTGGGACAAG CAGGAAACAGCTATGACCTTCCACACCAGAAATCATATCCTC

58


TGTAAAACGACGGCCAGTAGTTCTCCATCCAAGGACTCTACTAAAC CAGGAAACAGCTATGACCGCCAATGCATCCAATTCCAC

58


TGTAAAACGACGGCCAGTAACAAACAGCATTCCTTCCAGC CAGGAAACAGCTATGACCCAGACCTGTAGTAGAGAATTACCAAATCTC

58


TGTAAAACGACGGCCAGTCTTCATTTTTATGGTGGTGTCATG CAGGAAACAGCTATGACCAACACATCAACCTGCTTTTAAACTC

58


TGTAAAACGACGGCCAGTATCACTGAATTTCCTAGACCCTATGG CAGGAAACAGCTATGACCAAGTACTGAAATGCAGGACTGACAAC

58


TGTAAAACGACGGCCAGTTTCATCAGTGTCAACGGATTTC CAGGAAACAGCTATGACCGTCCATCATTTCTGCAAGAGAG

58


TGTAAAACGACGGCCAGTAGCATATGAACCAAAACTAACACC CAGGAAACAGCTATGACCCATAGTTAGGGATTCAGAAAAGAGC

58


TGTAAAACGACGGCCAGTGCGTCTTAGGTTATTTGTTTAGTTTGC CAGGAAACAGCTATGACCGCCTCTTTCAATGCAGTAATATGATTCT

58


TGTAAAACGACGGCCAGTTCCTCCATAGCTCAAAGGGAATAA CAGGAAACAGCTATGACCGGCATGGGAAGAGATTAATGACA

58


TGTAAAACGACGGCCAGTCACATCATAACACTTTTCCACCAGT CAGGAAACAGCTATGACCTTCTGAGATTTTCCCTGACAAGTAGAA

58



IV


TGTAAAACGACGGCCAGTGCTTTATCTTCCAGGTTCTGTAGCTAG CAGGAAACAGCTATGACCAGTGCTGGTCTTATTATTGAGTGAGC

58


TGTAAAACGACGGCCAGTGGGAAACAGAGAAAAGAATAGATGCT CAGGAAACAGCTATGACCGAATGTACTGAAAACCAGGTATCTTCAT

58


TGTAAAACGACGGCCAGTGGAAAATAGAGCAGCTTACCTTAGATACTG CAGGAAACAGCTATGACCCATGCTGTTCTGATGTAACTGCC

58


TGTAAAACGACGGCCAGTCACAATAAGCACTAAGATCATGCCTAG CAGGAAACAGCTATGACCGGGTGACAGAGCGAGACTCC

58


TGTAAAACGACGGCCAGTCAATCCATTTCTTTCTGTTCTGACC CAGGAAACAGCTATGACCAAGATCAAGAAACCAAGTGTCAGGA

58


TGTAAAACGACGGCCAGTGGAGAAAGTGGAGTTTGTGAAATAGAC CAGGAAACAGCTATGACCAAAGAATTCATCATCTGTGAATCAGAGTAC

58


TGTAAAACGACGGCCAGTGACTTTATGGGACAATATCACAGGAA CAGGAAACAGCTATGACCCCCTGTTAGACTAAATAATTCATTCTCAGG

58


TGTAAAACGACGGCCAGTCAAAAAAGGCAAACTTCAGCTATCA CAGGAAACAGCTATGACCTAAGACAAATCAACCCAAAGCATTC

58


TGTAAAACGACGGCCAGTCATGATGTTGTCCTTGATTAAGTAGTTG CAGGAAACAGCTATGACCCATAGTCCACAGTATCAGAGTTCATCCT

58


TGTAAAACGACGGCCAGTCACACCTTCTCAGTTTAGCACATTAA CAGGAAACAGCTATGACCGCCGGAGAACCTAAAAACAAA

58


TGTAAAACGACGGCCAGTAACGAAAGGCTCCAAAGTATGG CAGGAAACAGCTATGACCGAACACCTCGTTGCATATCAAGC

58


TGTAAAACGACGGCCAGTCCCAGTATTCTAGCAAGGGTAAAG CAGGAAACAGCTATGACCCCCATGTGGTTCCTATCATCTAC

58


TGTAAAACGACGGCCAGTTCTAGTCTTGTGTCCAGGGCTTAAC CAGGAAACAGCTATGACCTATTACCACTGAGGAATTGGTTGG

58


TGTAAAACGACGGCCAGTCCTGGCAGTTTGTGTTTTGATTAG CAGGAAACAGCTATGACCGGCAGCAGTACGAAATGTCAAG

58


TGTAAAACGACGGCCAGTACGTTGGTAAATGGTTGTGTTACTG CAGGAAACAGCTATGACCAAAACCACAACCAGGCAACTG

58


TGTAAAACGACGGCCAGTTGGAATAGGGAGAAAGGGAAGG CAGGAAACAGCTATGACCGCAGAGCCAGAGGCAAAAATG

58


TGTAAAACGACGGCCAGTCATGGCCTCTTGTTCTTTATCTTG CAGGAAACAGCTATGACCAGTTGATCTGTCAGTGAGAATGTGG

58


TGTAAAACGACGGCCAGTCAAAATGCCCTATTTCTGCCC CAGGAAACAGCTATGACCCACACCTCATAATTTCCTCGAGTCTAG

58


TGTAAAACGACGGCCAGTGGGTGACAGTGAGACTCCATCTC CAGGAAACAGCTATGACCGAGCAGCACTGTTTTCAGAATTTTC

58


TGTAAAACGACGGCCAGTACAGACTGATACTTTGAATGCCACTG CAGGAAACAGCTATGACCCCTACATTCTATACACCAAATCCTACTGC

58

a.2 Primers used from Krantz et al., 2004



V


Ta


g.NIPBL-10A-F g.NIPBL-10A-R

TTGCATTTGCATTTTACTCCA GTGTCTCAGGATGGTTTTCTGG

58

g.NIPBL-10B-F g.NIPBL-10B-R

TACGGGAAATGGGTCAAGGC AGGCTCAACTATGGTGCTCTCG

55

g.NIPBL-10C-F g.NIPBL-10C-R

TGAGAGCAGAACAACTGAATGC TGGCTTTCCAGGAATCCCTCC

55

g.NIPBL-10D-F g.NIPBL-10D-R

AGGTGAGAGCCGCCCTGAAACTC CACGAGGACTGTCAGGTCTTGA

55

g.NIPBL-10E-F g.NIPBL-10E-R

TGAATCAGGGGACTCAAGGG AGGGAACTTCTTGATTTGTCCTC

55

g.NIPBL-10F-F g.NIPBL-10F-R

AGGAGCTAAGCCTGTAGTTGTG CTTGAGTAGTGGGTGGGGAAGA

55


CTCTAGGTAAGGCCACCAGCAT TAGACCTCAGCATAAGGACTGC

55


CAGATTAAGAACCATTGAGCC GCAGTAATCATAACCCAAGAG

58


AGTGTGAGAATGCTTTATGTT ATTATGAATGTGGGCAGAGCA

55


ATGAAGCTAGCCTCAGAATGT CAAAATTTCCCCTTCACTTCTGA

58


GTGAGGTGAAAGTGCCCTGTA TCCCAAGTCAAGTATTGCCCAG

52


CAAGCTGTTGAATGGAGCATAC CATGAGCCACCACACCCAGC

58


TCCAAATACGTTGTTTCCATAG TCAATGTGAAGGAGATAGTTAT

58


CCACACCAAACTACTGCCATAG CATTTTACGTAATACGCTGCG

58

g.NIPBL-47A-F g.NIPBL-47A-R

GTCACGGTGCGTCTCATTGC TAGTGTCTACCCAAGGCACCA

58

g.NIPBL-47B-F g.NIPBL-47B-R

GGCTTCAGTGTTCAGTGGATG TTTGCCCAACATTTCCTTC

58

g.NIPBL-47C-F g.NIPBL-47C-R

TGAAGAGTAAGTGGAACCTGG GCTAAAGAAAGCCATCCGC

58



VI

b. SMC1A (Assession Number NM_006306)


Ta


g.SMC1A-1F g.SMC1A-1R

GGCCTGTCCTACTGCCGC CCGCGACGTTTCAGGTTAC

58


GACTTTCCCTCTACTGCCTCGTC

CCTCCTTCCTGTCCCAATCAA 58


TTGGGACAGGAAGGAGGCC GGGATGGGACATGGTCTGG

58


TGAACTGGGTTGGCTGGG TGGTCCCTCAGAGCCAAGAG

58


GCCTTATCGATGCACCCC GGTTTCCCACAAGTTGCAGG

58


GAAAAGCCTGGGAACCCTC CCACTGACAGCATCTCCTTCTC

58

g.SMC1A-6bF g.SMC1A-6bR

AGAAGCGGCCTCAGTACATCAA TCAGCCCAGGGATCTCCTC

58


ATCCCTGGGCTGACATCTG TCTGTGGACCATCATCCCTAATC

58


CCCATGGTCTTGGCGTAAACT GGCTGGTGAGAAAATACTCCTGG

58


TCTGAGCCAGGAGTATTTTCTCAC GTGTAAGTGCTCAGGAAATGAGTTC

58


GCTTCATACTTCAGTTTTCCCTTCC CCTTGAACACTGGCCTGACC

58


TTGTGTTCTGCCACTGACTGAG CTGAGCCTGTCCAGCTCCAG

58


CCCCTCTTCTCTCTTCCCCC GGACTCCCACTGCTAAAAACAGC

58


GCCACCACCTTACCACCTAGC GCCTTTGGAGAACAGGGCTG

58


TCCAAGGAGCCTGGTCTGC TGCCCTGACACACACGCTG

58


GACAGACTAAAAGGGCCGATG TGACCCTTGCCTATGACAGC

58


GAGATTATGAGTGATTCTCCCCCC GGACATTATCCTTCTGTCTGTCGTATT

58


GGGAAAGTCTTCACCTATTGTCCTG GGGTCATTTCCCTAAAAAAGGTCC

58


TGGGTCTCACCTAGCCCCTG TTCATCGCCCACTCCCAC

58

g.SMCA1-19F g.SMCA1-19R

CTGCTCCACTTGACTCTCCGT GAAGATAGTCCCACTCCCACC

58


CAGGCCCTTCTTGACATTGC TTGTCCTCATCGTAGGCCC

58


AGGTCCTTCTTGCCAGGGC GACAACTGAGGATGGTAGCAGC

58



VII


TCCCTATCTCCATCCAGTCTCAG AGTCTCTTCGTCAACTGCCCTAG

58


TGGCCTGTTTGATGGGCC AACCCAACCCCGACCTGG

58


TGCCTTCTGGTTGTGGCTG CAGGGAGTAGGACTGGCTCC

58


AGGGAGGAGGGTTTGAGGC CAGAGATTGGGAGAGGGACAG

58